{ "Video Omni": "Omni-Video\nCreate Task\nPOST\n/v1/videos/omni-video\ncURL\nCopy\nCollapse\ncurl --request POST \\\n --url https://api-singapore.klingai.com/v1/videos/omni-video \\\n --header 'Authorization: Bearer ' \\\n --header 'Content-Type: application/json' \\\n --data '{\n \"model_name\": \"kling-video-o1\",\n \"prompt\": \"Make the person in <<>> wave to the camera\",\n \"image_list\": [\n {\n \"image_url\": \"https://p2-kling.klingai.com/kcdn/cdn-kcdn112452/kling-qa-test/multi-1.png\"\n }\n ],\n \"duration\": \"5\",\n \"mode\": \"pro\",\n \"aspect_ratio\": \"16:9\",\n \"callback_url\": \"\",\n \"external_task_id\": \"\"\n}'\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes; Specific definitions can be found in \"Error Code\"\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system, is used for tracking requests and troubleshooting\n \"data\": {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_info\": { //Task creation parameters\n \"external_task_id\": \"string\" //User-defined task ID\n },\n \"task_status\": \"string\", // Task status, Enum values: submitted, processing, succeed, failed\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit: ms\n \"updated_at\": 1722769557708 //Task update time, Unix timestamp, unit: ms\n }\n}\n\nThe Omni model can achieve various capabilities through Prompt with elements, images, videos, and other content.\n\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nRequest Body\nmodel_name\nstring\nOptional\nDefault to kling-video-o1\n\nModel Name\n\nEnum values：\nkling-video-o1\nkling-v3-omni\nmulti_shot\nboolean\nOptional\nDefault to false\n\nWhether to generate multi-shot video.\n\nWhen true: the prompt parameter is invalid, and it does not support setting the start & end frames to generate videos\n\nWhen false: the shot_type and multi_prompt parameters are invalid.\n\nshot_type\nstring\nOptional\n\nStoryboard method.\n\nEnum values：\ncustomize\nintelligence\n\nWhen multi_shot is true, this parameter is required.\n\nprompt\nstring\nOptional\n\nText prompt words, which can include positive and negative descriptions.\n\nThe prompt words can be templated to meet different video generation needs\nMust not exceed 2,500 characters\n\nWhen the \"multi_shot\" parameter is set to false or when the \"multi_shot\" parameter is set to true and the \"shot-type\" parameter is set to intelligence, the current parameter must not be empty.\n\n💡\n\nThe Omni model can achieve various capabilities through Prompt with elements, images, videos, and other content:\n\nSpecify elements/images/videos using <<<>>> format, e.g.: <<>>, <<>>, <<>>\nFor detailed capabilities, see: KLING Omni Model User Guide, Kling VIDEO 3.0 Omni Model User Guide\n\nThe support range for different model versions and video modes varies. For details, see Capability Map\n\nmulti_prompt\narray\nOptional\n\nInformation about each storyboard, such as prompts and duration.\n\nDefine shot sequence number, corresponding prompt and duration via index, prompt, and duration parameters.\n\nSupports up to 6 storyboards, minimum 1.\nMax length per storyboard content: 512.\nEach storyboard duration must not exceed total task duration and must be ≥ 1.\nSum of all storyboard durations must equal total task duration.\n\nWhen multi_shot is true and shot_type is customize, this parameter is required. Format:\n\n\"multi_prompt\":[\n { \"index\": int, \"prompt\": \"string\", \"duration\": \"5\" },\n { \"index\": int, \"prompt\": \"string\", \"duration\": \"5\" }\n]\nimage_list\narray\nOptional\n\nReference Image List, including element, scene, style reference images.\n\nIncluding reference images of the element, scene, style, etc., it can also be used as the start or end frame to generate videos; When generating a video as the start or end frame:\nDefine whether the image is in the first and last frames using the type parameter: first_frame is the start frame, end_frame is the end frame.\nCurrently does not support only the end frame, which means that when there is a end frame image, there must be a first frame image.\nIf the image is not the start & end frame, do not configure the type parameter.\nWhen generating a video using the first frame or the first and last frames, video editing functions cannot be used.\n\nLoad with key:value, details as follows:\n\n\"image_list\":[\n { \"image_url\": \"image_url\"},\n { \"image_url\": \"image_url\" }\n]\n\nImage Requirements:\n\nSupports Base64 encoding or image URL (ensure accessibility)\nFormats: .jpg / .jpeg / .png\nFile size: ≤10MB\nDimensions: min 300px, aspect ratio 1:2.5 ~ 2.5:1\n\nQuantity Limits:\n\nWhen there are reference video, the sum of the amount of reference image and the amount of reference element shall not exceed 4.\nWhen there is no reference video, the sum of the amount of reference image and the amount of reference element shall not exceed 7.\nWhen using the O1 model, setting the start & end frame is not supported when there are more than two images in the array.\nimage_url parameter value must not be empty\n▾\nHide child attributes\nimage_url\nstring\nRequired\n\nImage URL or Base64\n\ntype\nstring\nOptional\n\nFrame type: first_frame or end_frame\n\nEnum values：\nfirst_frame\nend_frame\nelement_list\narray\nOptional\n\nReference Element List, based on element ID from element library.\n\nWhen generating a video using the first frame, a maximum of 3 elements are supported.\nWhen generating videos using the start & end frames, the kling-v3-omni model supports up to 3 elements, while the kling-video-o1 model does not support 1 element.\nWhen there is a reference video, the sum of the number of reference images and the number of reference subjects must not exceed 4, and the use of video subjects is not supported.\nWhen there is no reference video, the sum of the number of reference images and the number of reference subjects must not exceed 7.\n\nElements are categorized into video character elements and Multi-Image Elements, with different scopes. See Kling Element Library User Guide.\n\nLoad with key:value format as follows:\n\"element_list\":[\n { \"element_id\": long },\n { \"element_id\": long }\n]\n\nThe support range for different model versions and video modes varies. For details, see Capability Map\n\n▾\nHide child attributes\nelement_id\nlong\nRequired\n\nElement ID from element library\n\nvideo_list\narray\nOptional\n\nReference Video, obtained via URL.\n\nVideo Types:\n\nCan be used as feature reference video OR base video for editing (default: base)\nUse refer_type parameter: feature for feature reference, base for video to be edited\nWhen reference video is base type, first/end frames cannot be defined\nUse keep_original_sound to keep original sound: yes to keep, no to discard (also applies to feature type)\n\nWhen there is a reference video, the value of the sound parameter can only be off.\n\nLoad with key:value format as follows:\n\"video_list\":[\n { \"video_url\": \"video_url\", \"refer_type\": \"base\", \"keep_original_sound\": \"yes\" }\n]\n\nVideo Requirements:\n\nFormat: MP4/MOV only\nDuration: ≥3 seconds, upper limit depends on model version (see Capability Map)\nResolution: 720px-2160px (width and height)\nFrame rate: 24-60fps (output will be 24fps)\nMax 1 video, size ≤200MB\nvideo_url parameter value must not be empty\n\nThe support range for different model versions and video modes varies. For details, see Capability Map\n\n▾\nHide child attributes\nvideo_url\nstring\nRequired\n\nVideo URL\n\nrefer_type\nstring\nOptional\nDefault to base\n\nReference type: feature (feature reference) or base (video to be edited)\n\nEnum values：\nfeature\nbase\nbase - Transformation (Video Edit):\nEdit video content - add/remove/modify elements, change shot composition, angles, styles, colors, weather, etc.\nfeature - Video Reference:\nReference video content to generate next/previous shot, or reference video style/camera movement.\nkeep_original_sound\nstring\nOptional\n\nKeep original sound: yes to keep, no to discard\n\nEnum values：\nyes\nno\nsound\nstring\nOptional\nDefault to off\n\nWhether to generate sound simultaneously when generating videos.\n\nEnum values：\non\noff\n\nThe support range for different model versions and video modes varies. For details, see Capability Map\n\nmode\nstring\nOptional\nDefault to pro\n\nVideo generation mode\n\nEnum values：\nstd\npro\nstd: Standard mode, cost-effective\npro: Professional mode (high quality), better video quality output\n\nThe support range for different model versions and video modes varies. For details, see Capability Map\n\naspect_ratio\nstring\nOptional\n\nAspect ratio of generated video frame (width:height)\n\nEnum values：\n16:9\n9:16\n1:1\nThis parameter is required when NOT using first-frame reference or video editing function.\nduration\nstring\nOptional\nDefault to 5\n\nVideo duration in seconds\n\nEnum values：\n3\n4\n5\n6\n7\n8\n9\n10\n11\n12\n13\n14\n15\nWhen using the video editing function (\"refer_type\": \"base\"), the output result is the same as the duration of the incoming video, and the current parameter is invalid. Calculate billing by rounding the input video duration to the nearest integer.\n\nThe support range for different model versions and video modes varies. For details, see Capability Map\n\nwatermark_info\nobject\nOptional\n\nWhether to generate watermarked results simultaneously\n\nDefined by the enabled parameter, format:\n \"watermark_info\": { \"enabled\": boolean } \ntrue: generate watermarked result, false: do not generate\nCustom watermarks are not currently supported\ncallback_url\nstring\nOptional\n\nCallback notification URL for task result. If configured, server will actively notify when task status changes.\n\nFor specific message schema, see Callback Protocol\nexternal_task_id\nstring\nOptional\n\nCustom task ID defined by user.\n\nWill not overwrite system-generated task ID, but supports querying task by this ID.\nMust be unique within a single user account.\nScenario invocation examples\n\nThe following is an example of scene code. For more effects and prompt words, please refer to: Kling Omni Model Example\n\nImage/Element Reference\nImage/Element Reference：Supports reference images/elements, including characters, items, backgrounds, and more, to generate with more creativity and consistency.\ncURL\nCopy\nCollapse\ncurl --location 'https://api-singapore.klingai.com/v1/videos/omni-video' \\\n--header 'Authorization: Bearer ' \\\n--header 'Content-Type: application/json' \\\n--data '{\n \"model_name\": \"kling-video-o1\",\n \"prompt\": \"<<>> strolling through the streets of Tokyo, encountered <<>> and <<>>, and jumped into the arms of <<>>. The video style matches that of <<>>\",\n \"image_list\": [\n {\n \t\"image_url\": \"xxxxx\"\n },\n {\n \t\"image_url\": \"xxxxx\"\n }\n ],\n \"element_list\": [\n {\n \t\"element_id\": long\n },\n {\n \t\"element_id\": long\n }\n ],\n \"mode\": \"pro\",\n \"aspect_ratio\": \"1:1\",\n \"duration\": \"7\"\n}'\nTransformation\nInput-based Modification: Supports Inpainting/outpainting, or changing shot compositions or angles. It also supports localized or full-scale adjustments, such as modifying/swapping subjects, backgrounds, partial areas, styles, colors, weather, and more.\ncURL\nCopy\nCollapse\ncurl --location 'https://api-singapore.klingai.com/v1/videos/omni-video' \\\n--header 'Authorization: Bearer ' \\\n--header 'Content-Type: application/json' \\\n--data '{\n \"model_name\": \"kling-video-o1\",\n \"prompt\": \"Put the crown from <<>> on the girl in blue from <<>>.\",\n \"image_list\": [\n {\n \t\"image_url\": \"xxx\"\n }\n ],\n \"video_list\": [\n {\n \"video_url\":\"xxxxxxxx\",\n \"refer_type\":\"base\",\n \"keep_original_sound\":\"yes\"\n }\n ],\n \"mode\": \"pro\"\n}'\nVideo Reference\nVideo Reference: Supports using reference video content to generate previous or next shots within the same context or set. It can also reference video actions or camera movements for generation.\ncURL\nCopy\nCollapse\ncurl --location 'http", "Text to Video": "Motion Control\nMulti-elements to video\nExtend Video\nLip Sync\nAvatar\nText to Audio\nVideo to Audio\nText to Speech\nVoice Clone\nImage Recognize\nElement\nEffects\nEffect Templates\nNEW\nVideo Effects\nImage Generation\nModels\nImage Omni\nImage Generation\nReference to Image\nExtend Image\nAI Multi-Shot\nVirtual Try-On\nOthers\nQuery user info\nPricing\nBilling Info\nPrepaid Resource Packs\nProtocols\nPrivacy Policy of API Service\nTerms of API Service\nAPI Service Level Agreement\nText to Video\nCreate Task\nPOST\n/v1/videos/text2video\ncURL\nCopy\nCollapse\ncurl --request POST \\\n --url https://api-singapore.klingai.com/v1/videos/text2video \\\n --header 'Authorization: Bearer ' \\\n --header 'Content-Type: application/json' \\\n --data '{\n \"model_name\": \"kling-v2-6\",\n \"prompt\": \"A cute little rabbit wearing glasses, sitting at a table, reading a newspaper, with a cup of cappuccino on the table\",\n \"negative_prompt\": \"\",\n \"duration\": \"5\",\n \"mode\": \"pro\",\n \"sound\": \"on\",\n \"aspect_ratio\": \"1:1\",\n \"callback_url\": \"\",\n \"external_task_id\": \"\"\n }'\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes; Specific definitions can be found in \"Error Code\"\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system\n \"data\": {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_info\": { // Task creation parameters\n \"external_task_id\": \"string\" // Customer-defined task ID\n },\n \"task_status\": \"string\", // Task status, Enum values: submitted, processing, succeed, failed\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit ms\n \"updated_at\": 1722769557708 // Task update time, Unix timestamp, unit ms\n }\n}\n💡\n\nPlease note that in order to maintain naming consistency, the original model field has been changed to model_name, so in the future, please use this field to specify the version of the model that needs to be called.\nAt the same time, we keep the behavior forward-compatible. If you continue to use the original model field, it will not have any impact on the interface call, there will not be any exception, which is equivalent to the default behavior when model_name is empty (i.e., call the V1 model).\n\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authorization\n\nRequest Body\nmodel_name\nstring\nOptional\nDefault to kling-v1\n\nModel Name\n\nEnum values：\nkling-v1\nkling-v1-6\nkling-v2-master\nkling-v2-1-master\nkling-v2-5-turbo\nkling-v2-6\nkling-v3\nmulti_shot\nboolean\nOptional\nDefault to false\n\nWhether to generate multi-shot video\n\nWhen true: the prompt parameter is invalid, and the first/end frame generation is not supported.\n\nWhen false: the shot_type and multi_prompt parameters are invalid\n\nshot_type\nstring\nOptional\n\nStoryboard method\n\nEnum values：\ncustomize\nintelligence\n\nWhen multi_shot is true, this parameter is required\n\nprompt\nstring\nOptional\n\nPositive text prompt\n\n💡\n\nThe Omni model can achieve various capabilities through Prompt with elements, images, videos, and other content:\n\nSpecify elements/images/videos using <<<>>> format, e.g.: <<>>, <<>>, <<>>\nFor detailed capabilities, see: KLING Omni Model User Guide, Kling VIDEO 3.0 Omni Model User Guide\nCannot exceed 2500 characters\nUse <<>> to specify voice, same sequence as voice_list. Up to 2 voices; when specifying voice, sound must be on. Simpler grammar is better. For example: The man <<>> said, \"Hello.\".\nWhen voice_list is not empty and prompt references voice ID, task is billed as \"with voice generation\".\nWhen multi_shot is false or shot_type is intelligence, this parameter must not be empty.\n\nThe support range for different model versions and video modes varies. For details, see Capability Map\n\nmulti_prompt\narray\nOptional\n\nEach storyboard cue can include both positive and negative descriptions\n\nDefine the shot sequence number, corresponding prompt word, and duration through the index, prompt, and duration parameters, where:\n\nSupports up to 6 storyboards, with a minimum of 1 storyboard.\nThe maximum length of the prompt for each storyboard 512 characters.\nThe duration of each storyboard should not exceed the total duration, but should not be less than 1.\nThe sum of the durations of all storyboards equals the total duration of the current task.\n\nLoad with key:value format as follows:\n\n\"multi_prompt\":[\n{\"index\":int,\"prompt\":\"string\",\"duration\":\"5\"},\n{\"index\":int,\"prompt\":\"string\",\"duration\":\"5\"}\n]\n\nWhen multi_shot is true and shot_type is customize, this parameter is required.\n\nnegative_prompt\nstring\nOptional\n\nNegative text prompt\n\nCannot exceed 2500 characters\nIt is recommended to supplement negative prompt via negative sentences within positive prompts\nsound\nstring\nOptional\nDefault to off\n\nIs sound generated simultaneously when generating videos\n\nEnum values：\non\noff\n\nThe support range for different model versions and video modes varies. For details, see Capability Map\n\ncfg_scale\nfloat\nOptional\nDefault to 0.5\n\nThe degree of freedom for generating video; the larger the value, the smaller the degree of freedom of the model\n\nValue range: [0, 1]\n\nkling-v2.x models do not support this parameter\n\nmode\nstring\nOptional\nDefault to std\n\nVideo generation mode\n\nEnum values：\nstd\npro\nstd: Standard mode, basic mode, cost-effective\npro: Expert mode (high quality), high performance mode, better video quality\n\nThe support range for different model versions and video modes varies. For details, see Capability Map\n\ncamera_control\nobject\nOptional\n\nTerms of controlling camera movement (if not specified, the model will intelligently match based on the input text/images)\n\nThe support range for different model versions and video modes varies. For details, see Capability Map\n\n▾\nHide child attributes\ntype\nstring\nOptional\n\nPredefined camera movements type\n\nEnum values：\nsimple\ndown_back\nforward_up\nright_turn_forward\nleft_turn_forward\nsimple: Simple camera movement, you can choose one of six options in \"config\"\ndown_back: Camera descends and moves backward ➡️ Pan down and zoom out. The config parameter must be set to \"None\" under this type.\nforward_up: Camera moves forward and tilts up ➡️ Zoom in and pan up. The config parameter must be set to \"None\" under this type.\nright_turn_forward: Rotate right then move forward ➡️ Rotate right and advance. The config parameter must be set to \"None\" under this type.\nleft_turn_forward: Rotate left then move forward ➡️ Rotate left and advance. The config parameter must be set to \"None\" under this type.\nconfig\nobject\nOptional\n\nContains 6 fields, used to specify the camera's movement or change in different directions\n\nRequired when type is simple, not required for other types\nChoose 1 of the following 6 parameters, only one can be non-zero, others must be 0\n▾\nHide child attributes\nhorizontal\nfloat\nOptional\n\nHorizontal, controls the camera's movement along the horizontal axis (translation along the x-axis)\n\nValue range: [-10, 10]\nNegative value indicates a translation to the left, positive value indicates a translation to the right\nvertical\nfloat\nOptional\n\nVertical, controls the camera's movement along the vertical axis (translation along the y-axis)\n\nValue range: [-10, 10]\nNegative value indicates a downward translation, positive value indicates an upward translation\npan\nfloat\nOptional\n\nPan, controls the camera's rotation in the horizontal plane (rotation around the y-axis)\n\nValue range: [-10, 10]\nNegative value indicates rotation to the left around y-axis, positive value indicates rotation to the right around y-axis\ntilt\nfloat\nOptional\n\nTilt, controls the camera's rotation in the vertical plane (rotation around the x-axis)\n\nValue range: [-10, 10]\nNegative value indicates rotation down around x-axis, positive value indicates rotation up around x-axis\nroll\nfloat\nOptional\n\nRoll, controls the camera's roll (rotation around the z-axis)\n\nValue range: [-10, 10]\nNegative value indicates counterclockwise rotation around z-axis, positive value indicates clockwise rotation around z-axis\nzoom\nfloat\nOptional\n\nZoom, controls the camera's focal length change, affecting the distance of the field of view\n\nValue range: [-10, 10]\nNegative value indicates longer focal length, smaller field of view; positive value indicates shorter focal length, larger field of view\naspect_ratio\nstring\nOptional\nDefault to 16:9\n\nThe aspect ratio of the generated video frame (width:height)\n\nEnum values：\n16:9\n9:16\n1:1\nduration\nstring\nOptional\nDefault to 5\n\nVideo Length, unit: s (seconds)\n\nEnum values：\n3\n4\n5\n6\n7\n8\n9\n10\n11\n12\n13\n14\n15\n\nThe support range for different model versions and video modes varies. For details, see Capability Map\n\nwatermark_info\nobject\nOptional\n\nWhether to generate watermarked results simultaneously\n\nDefined by the enabled parameter, format:\n \"watermark_info\": { \"enabled\": boolean } \ntrue: generate watermarked result, false: do not generate\nCustom watermarks are not currently supported\ncallback_url\nstring\nOptional\n\nCallback notification URL for this task result. If configured, the server will actively notify when the task status changes\n\nFor specific message schema, see Callback Protocol\nexternal_task_id\nstring\nOptional\n\nCustomized Task ID\n\nUsers can provide a customized task ID, which will not overwrite the system-generated task ID but can be used for task queries\nMust be unique within a single user account\nMore Scene Invocation Examples\nText To Video with Multiple Shot\ncurl --location 'https://xxx/v1/videos/text2video' \\\n--header 'Authorization: Bearer xxx' \\\n--header 'Content-Type: application/json' \\\n--data '{\n \"model_name\": \"kling-v3\",\n \"prompt\": \"\",\n \"multi_prompt\": [\n {\n \"index\": 1,\n \"prompt\": \"Two friends talking under a streetlight at night. Warm glow, casual poses, no dialogue.\",\n \"duration\": \"2\"\n },\n {\n \"index\": 2,\n \"prompt\": \"A runner sprinting through a forest, leaves flying. Low-angle shot, focus on movement.\",\n \"duration\": \"3\"\n },\n {\n \"index\": 3,\n \"prompt\": \"A woman hugging a cat, smiling. Soft sunlight, cozy home setting, emphasize warmth.\",\n \"duration\": \"3\"\n },\n {\n \"index\": 4,\n \"prompt\": \"A door creaking open, shadowy hallway. Dark tones, minimal details, eerie mood.\",\n \"duration\": \"3\"\n },\n {\n \"index\": 5,\n \"prompt\": \"A man slipping on a banana peel, shocked expression. Exaggerated pose, bright colors.\",\n \"duration\": \"3\"\n },\n {\n \"index\": 6,\n \"prompt\": \"A sunset over mountains, small figure walking away. Wide angle, peaceful atmosphere.\",\n \"duration\": \"1\"\n }\n ],\n \"multi_shot\": true,\n \"shot_type\": \"customize\",\n \"duration\": \"15\",\n \"mode\": \"pro\",\n \"sound\": \"on\",\n \"aspect_ratio\": \"9:16\",\n \"callback_url\": \"\",\n \"external_task_id\": \"\"\n}'\nQuery Task (Single)\nGET\n/v1/videos/text2video/{id}\ncURL\nCopy\nCollapse\ncurl --request GET \\\n --url https://api-singapore.klingai.com/v1/videos/text2video/{task_id} \\\n --header 'Authorization: Bearer '\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes; Specific definitions can be found in \"Error Code\"\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system, is used to track requests and troubleshoot problems\n \"data\": {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_status\": \"string\", // Task status, Enum values: submitted, processing, succeed, failed\n \"task_status_msg\": \"string\", // Task status information, displaying the failure reason when the task fails (such as triggering the content risk control of the platform, etc.)\n \"task_info\": { // Task creation parameters\n \"external_task_id\": \"string\" // Customer-defined task ID\n },\n \"task", "Image to Video": "Motion Control\nMulti-elements to video\nExtend Video\nLip Sync\nAvatar\nText to Audio\nVideo to Audio\nText to Speech\nVoice Clone\nImage Recognize\nElement\nEffects\nEffect Templates\nNEW\nVideo Effects\nImage Generation\nModels\nImage Omni\nImage Generation\nReference to Image\nExtend Image\nAI Multi-Shot\nVirtual Try-On\nOthers\nQuery user info\nPricing\nBilling Info\nPrepaid Resource Packs\nProtocols\nPrivacy Policy of API Service\nTerms of API Service\nAPI Service Level Agreement\nImage to Video\nCreate Task\nPOST\n/v1/videos/image2video\ncURL\nCopy\nCollapse\ncurl --location --request POST 'https://api-singapore.klingai.com/v1/videos/image2video' \\\n--header 'Authorization: Bearer ' \\\n--header 'Content-Type: application/json' \\\n--data-raw '{\n \"model_name\": \"kling-v2-6\",\n \"image\": \"https://p2-kling.klingai.com/kcdn/cdn-kcdn112452/kling-qa-test/multi-2.png\",\n \"image_tail\": \"https://p2-kling.klingai.com/kcdn/cdn-kcdn112452/kling-qa-test/multi-1.png\",\n \"prompt\": \"Camera zooms out, the girl smiles\",\n \"negative_prompt\": \"\",\n \"duration\": \"5\",\n \"mode\": \"pro\",\n \"sound\": \"off\",\n \"callback_url\": \"\",\n \"external_task_id\": \"\"\n}'\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes; Specific definitions can be found in \"Error Code\"\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system\n \"data\": {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_info\": { // Task creation parameters\n \"external_task_id\": \"string\" // Customer-defined task ID\n },\n \"task_status\": \"string\", // Task status, Enum values: submitted, processing, succeed, failed\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit ms\n \"updated_at\": 1722769557708 // Task update time, Unix timestamp, unit ms\n }\n}\n💡\n\nPlease note that in order to maintain naming consistency, the original model field has been changed to model_name. Please use this field to specify the model version in the future.\nWe maintain backward compatibility. If you continue using the original model field, it will not affect API calls and will be equivalent to the default behavior when model_name is empty (i.e., calling the V1 model).\n\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nRequest Body\nmodel_name\nstring\nOptional\nDefault to kling-v1\n\nModel Name\n\nEnum values：\nkling-v1\nkling-v1-5\nkling-v1-6\nkling-v2-master\nkling-v2-1\nkling-v2-1-master\nkling-v2-5-turbo\nkling-v2-6\nkling-v3\nimage\nstring\nOptional\n\nReference Image\n\nSupports image Base64 encoding or image URL (ensure accessibility)\nImportant: When using Base64, do NOT add any prefix like data:image/png;base64,. Submit only the raw Base64 string.\nCorrect Base64 format:\niVBORw0KGgoAAAANSUhEUgAAAAUA...\nIncorrect Base64 format (with data: prefix):\ndata:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAUA...\nSupported image formats: .jpg / .jpeg / .png\nFile size: ≤10MB, dimensions: min 300px, aspect ratio: 1:2.5 ~ 2.5:1\nAt least one of image or image_tail must be provided; both cannot be empty\n\nSupport varies by model version and video mode. See Capability Map for details.\n\nimage_tail\nstring\nOptional\n\nReference Image - End frame control\n\nSupports image Base64 encoding or image URL (ensure accessibility)\nImportant: When using Base64, do NOT add any prefix like data:image/png;base64,. Submit only the raw Base64 string.\nSupported image formats: .jpg / .jpeg / .png\nFile size: ≤10MB, dimensions: min 300px\nAt least one of image or image_tail must be provided; both cannot be empty\nimage_tail, dynamic_masks/static_mask, and camera_control are mutually exclusive - only one can be used at a time\n\nSupport varies by model version and video mode. See Capability Map for details.\n\nmulti_shot\nboolean\nOptional\nDefault to false\n\nWhether to generate multi-shot video\n\nWhen true: the prompt parameter is invalid, and the first/end frame generation is not supported.\n\nWhen false: the shot_type and multi_prompt parameters are invalid\n\nshot_type\nstring\nOptional\n\nStoryboard method\n\nEnum values：\ncustomize\nintelligence\n\nWhen multi_shot is true, this parameter is required\n\nprompt\nstring\nOptional\n\nPositive text prompt\n\n💡\n\nThe Omni model can achieve various capabilities through Prompt with elements, images, videos, and other content:\n\nSpecify elements/images/videos using <<<>>> format, e.g.: <<>>, <<>>, <<>>\nFor detailed capabilities, see: KLING Omni Model User Guide, Kling VIDEO 3.0 Omni Model User Guide\nCannot exceed 2500 characters\nWhen multi_shot is false or shot_type is intelligence, this parameter must not be empty.\nUse <<>> to specify voice, with the sequence matching the voice_list parameter order\nA video generation task can reference up to 2 voices; when specifying a voice, the sound parameter must be \"on\"\nThe simpler the syntax structure, the better. Example: The man<<>> said: \"Hello\"\nWhen voice_list is not empty and prompt references voice ID, the task will be billed as \"with specified voice\"\n\nSupport varies by model version and video mode. See Capability Map for details.\n\nmulti_prompt\narray\nOptional\n\nInformation about each storyboard, such as prompts and duration\n\nDefine the shot sequence number, corresponding prompt word, and duration through the index, prompt, and duration parameters, where:\n\nSupports up to 6 storyboards, with a minimum of 1 storyboard.\nThe maximum length of the prompt for each storyboard 512 characters.\nThe duration of each storyboard should not exceed the total duration, but should not be less than 1.\nThe sum of the durations of all storyboards equals the total duration of the current task.\n\nLoad with key:value format as follows:\n\n\"multi_prompt\":[\n{\"index\":int,\"prompt\":\"string\",\"duration\":\"5\"},\n{\"index\":int,\"prompt\":\"string\",\"duration\":\"5\"}\n]\n\nWhen multi_shot is true and shot_type is customize, this parameter is required.\n\nnegative_prompt\nstring\nOptional\n\nNegative text prompt\n\nCannot exceed 2500 characters\nIt is recommended to supplement negative prompt via negative sentences within positive prompts\nelement_list\narray\nOptional\n\nReference Element List, based on element ID from element library\n\nSupports up to 3 reference elements\n\nThe elements are categorized into video customization element (named as Video Character Elements) and image customization elements (named as Multi-Image Elements), each with distinct scopes of application. Please exercise caution in distinguishing between them. See Kling Element Library User Guide.\n\nLoad with key:value format as follows:\n\"element_list\":[\n { \"element_id\": long },\n { \"element_id\": long }\n]\n\nSupport varies by model version and video mode. See Capability Map for details.\n\n▾\nHide child attributes\nelement_id\nlong\nRequired\n\nElement ID from element library\n\nvoice_list\narray\nOptional\n\nList of voices referenced when generating videos\n\nA video generation task can reference up to 2 voices\nWhen voice_list is not empty and prompt references voice ID, the task will be billed as \"with specified voice\"\nvoice_id is returned through the voice customization API, or use system preset voices. See Custom Voices API; NOT the voice_id of Lip-Sync API\nelement_list and voice_list are mutually exclusive and cannot coexist\n\nExample:\n\n\"voice_list\":[\n {\"voice_id\":\"voice_id_1\"},\n {\"voice_id\":\"voice_id_2\"}\n]\n\nThe support range for different model versions and video modes varies. For details, see Capability Map\n\nsound\nstring\nOptional\nDefault to off\n\nWhether to generate sound when generating video\n\nEnum values：\non\noff\n\nThe support range for different model versions and video modes varies. For details, see Capability Map\n\ncfg_scale\nfloat\nOptional\nDefault to 0.5\n\nFlexibility in video generation; higher value means lower model flexibility and stronger relevance to user prompt\n\nValue range: [0, 1]\n\nkling-v2.x models do not support this parameter\n\nmode\nstring\nOptional\nDefault to std\n\nVideo generation mode\n\nEnum values：\nstd\npro\nstd: Standard Mode - basic mode, cost-effective\npro: Professional Mode (High Quality) - high performance mode, better video quality\n\nSupport varies by model version and video mode. See Capability Map for details.\n\nstatic_mask\nstring\nOptional\n\nStatic brush mask area (mask image created by user using motion brush)\n\nThe \"Motion Brush\" feature includes Dynamic Brush (dynamic_masks) and Static Brush (static_mask)\n\nSupports image Base64 encoding or image URL (same format requirements as image field)\nSupported image formats: .jpg / .jpeg / .png\nAspect ratio must match the input image (image field), otherwise task will fail\nResolution of static_mask and dynamic_masks.mask must be identical, otherwise task will fail\n\nSupport varies by model version and video mode. See Capability Map for details.\n\ndynamic_masks\narray\nOptional\n\nDynamic brush configuration list\n\nCan configure multiple groups (up to 6), each containing \"mask area\" and \"motion trajectory\" sequence\n\nSupport varies by model version and video mode. See Capability Map for details.\n\n▾\nHide child attributes\nmask\nstring\nRequired\n\nDynamic brush mask area (mask image created by user using motion brush)\n\nSupports image Base64 encoding or image URL (same format requirements as image field)\nSupported image formats: .jpg / .jpeg / .png\nAspect ratio must match the input image (image field), otherwise task will fail\nResolution of static_mask and dynamic_masks.mask must be identical, otherwise task will fail\ntrajectories\narray\nRequired\n\nMotion trajectory coordinate sequence\n\nFor 5s video, trajectory length ≤77, coordinate count range: [2, 77]\nCoordinate system uses bottom-left corner of image as origin\n\nNote 1: More coordinate points = more accurate trajectory. 2 points = straight line between them\n\nNote 2: Trajectory direction follows input order. First coordinate is start point, subsequent coordinates are connected sequentially\n\n▾\nHide child attributes\nx\nint\nRequired\n\nX coordinate of trajectory point (pixel coordinate with image bottom-left as origin)\n\ny\nint\nRequired\n\nY coordinate of trajectory point (pixel coordinate with image bottom-left as origin)\n\ncamera_control\nobject\nOptional\n\nCamera movement control protocol (if not specified, model will intelligently match based on input text/images)\n\nSupport varies by model version and video mode. See Capability Map for details.\n\n▾\nHide child attributes\ntype\nstring\nRequired\n\nPredefined camera movement type\n\nEnum values：\nsimple\ndown_back\nforward_up\nright_turn_forward\nleft_turn_forward\nsimple: Simple camera movement, can choose one of six options in \"config\"\ndown_back: Camera descends and moves backward ➡️ Pan down and zoom out. config parameter not required\nforward_up: Camera moves forward and tilts up ➡️ Zoom in and pan up. config parameter not required\nright_turn_forward: Rotate right then move forward ➡️ Right rotation advance. config parameter not required\nleft_turn_forward: Rotate left then move forward ➡️ Left rotation advance. config parameter not required\nconfig\nobject\nOptional\n\nContains 6 fields to specify camera movement in different directions\n\nRequired when type is \"simple\"; leave empty for other types\nChoose only one parameter to be non-zero; rest must be 0\n▾\nHide child attributes\nhorizontal\nfloat\nOptional\n\nHorizontal movement - camera translation along x-axis\n\nValue range: [-10, 10]. Negative = left, Positive = right\nvertical\nfloat\nOptional\n\nVertical movement - camera translation along y-axis\n\nValue range: [-10, 10]. Negative = down, Positive = up\npan\nfloat\nOptional\n\nHorizontal pan - camera rotation around y-axis\n\nValue range: [-10, 10]. Negative = rotate left, Positive = rotate right\ntilt\nfloat\nOptional\n\nVertical tilt - camera rotation around x-axis\n\nValue range: [-10, 10]. Negative = tilt down, Positive = tilt up\nroll\nfloat\nOptional\n\nRoll - camera rotation around z-axis\n\nValue range: [-10, 10]. Negative = counterclockwise, Positive = clockwise\nzoom\nfloat\nOptional\n\nZoom - controls camera focal length chang", "Reference to Video": "Multi-Image to Video\nCreate Task\nPOST\n/v1/videos/multi-image2video\ncURL\nCopy\nCollapse\ncurl --request POST \\\n --url https://api-singapore.klingai.com/v1/videos/multi-image2video \\\n --header 'Authorization: Bearer ' \\\n --header 'Content-Type: application/json' \\\n --data '{\n \"model_name\": \"kling-v1-6\",\n \"image_list\": [\n { \"image\": \"https://p1-kling.klingai.com/kcdn/cdn-kcdn112452/kling-qa-test/dog.png\" },\n { \"image\": \"https://p1-kling.klingai.com/kcdn/cdn-kcdn112452/kling-qa-test/dog_cloth.png\" }\n ],\n \"prompt\": \"A white Bichon Frise wearing a red Northeast-style floral cotton jacket, licking its paw\",\n \"negative_prompt\": \"\",\n \"mode\": \"pro\",\n \"duration\": \"5\",\n \"aspect_ratio\": \"16:9\",\n \"callback_url\": \"\",\n \"external_task_id\": \"\"\n }'\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error code; Specific definitions can be found in \"Error Code\"\n \"message\": \"string\", // Error message\n \"request_id\": \"string\", // Request ID, generated by the system, used to track requests and troubleshoot problems\n \"data\": {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_status\": \"string\", // Task status, Enum values: submitted, processing, succeed, failed\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit: ms\n \"updated_at\": 1722769557708 //Task update time, Unix timestamp, unit: ms\n }\n}\n\nGenerate video from multiple reference images (elements).\n\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nRequest Body\nmodel_name\nstring\nOptional\nDefault to kling-v1-6\n\nModel Name\n\nEnum values：\nkling-v1-6\nimage_list\narray\nRequired\n\nReference Image List\n\nSupport up to 4 images, load with key:value format as follows:\n\"image_list\":[\n { \"image\":\"image_url\" },\n { \"image\":\"image_url\" },\n { \"image\":\"image_url\" },\n { \"image\":\"image_url\" }\n]\nPlease directly upload the image with selected subject since there is no cropping logic on the API side\nSupports image input as either Base64-encoded string or URL (ensure the URL is publicly accessible)\nImportant: When using Base64, do NOT add any prefix like data:image/png;base64,. Submit only the raw Base64 string.\nCorrect Base64 format:\niVBORw0KGgoAAAANSUhEUgAAAAUA...\nIncorrect Base64 format (with data: prefix):\ndata:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAUA...\nSupported image formats: .jpg / .jpeg / .png\nImage file size must not exceed 10MB. Image dimensions must be at least 300px. Aspect ratio must be between 1:2.5 and 2.5:1\n▾\nHide child attributes\nimage\nstring\nRequired\n\nImage URL or Base64 string\n\nprompt\nstring\nRequired\n\nPositive text prompt\n\nCannot exceed 2500 characters\nnegative_prompt\nstring\nOptional\n\nNegative text prompt\n\nCannot exceed 2500 characters\nmode\nstring\nOptional\nDefault to std\n\nVideo generation mode\n\nEnum values：\nstd\npro\nstd: Standard Mode, which is cost-effective\npro: Professional Mode, generates higher quality video output\n\nDifferent model versions and video modes have different support ranges. For details, see Capability Map\n\nduration\nstring\nOptional\nDefault to 5\n\nVideo Length, unit: s (seconds)\n\nEnum values：\n5\n10\naspect_ratio\nstring\nOptional\nDefault to 16:9\n\nThe aspect ratio of the generated video frame (width:height)\n\nEnum values：\n16:9\n9:16\n1:1\nwatermark_info\nobject\nOptional\n\nWhether to generate watermarked results simultaneously\n\nDefined by the enabled parameter, format:\n \"watermark_info\": { \"enabled\": boolean } \ntrue: generate watermarked result, false: do not generate\nCustom watermarks are not currently supported\ncallback_url\nstring\nOptional\n\nThe callback notification address for the result of this task. If configured, the server will actively notify when the task status changes\n\nThe specific message schema of the notification can be found in Callback Protocol\nexternal_task_id\nstring\nOptional\n\nCustomized Task ID\n\nUser-defined task ID. It will not override the system-generated task ID, but supports querying tasks by this ID\nPlease note that it must be unique for each user\nQuery Task (Single)\nGET\n/v1/videos/multi-image2video/{id}\ncURL\nCopy\nCollapse\ncurl --request GET \\\n --url https://api-singapore.klingai.com/v1/videos/multi-image2video/{task_id} \\\n --header 'Authorization: Bearer '\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error code; Specific definitions can be found in \"Error Code\"\n \"message\": \"string\", // Error message\n \"request_id\": \"string\", // Request ID, generated by the system, used to track requests and troubleshoot problems\n \"data\": {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_status\": \"string\", // Task status, Enum values: submitted, processing, succeed, failed\n \"task_status_msg\": \"string\", // Task status message, displays failure reason when task fails (such as content moderation triggers)\n \"task_info\": { //Task creation parameters\n \"external_task_id\": \"string\" //User-defined task ID\n },\n \"task_result\": {\n \"videos\": [\n {\n \"id\": \"string\", // Generated video ID; globally unique\n \"url\": \"string\", // URL for generating videos (Please note that for security purposes, generated images/videos will be deleted after 30 days. Please save them promptly.)\n \"watermark_url\": \"string\", // Watermarked video download URL, anti-hotlinking format\n \"duration\": \"string\" //Total video duration, unit: s (seconds)\n }\n ]\n },\n \"watermark_info\": {\n \"enabled\": boolean\n },\n \"final_unit_deduction\": \"string\", // Final unit deduction for the task\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit: ms\n \"updated_at\": 1722769557708 //Task update time, Unix timestamp, unit: ms\n }\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nPath Parameters\ntask_id\nstring\nOptional\n\nTask ID for Multi-Image to Video\n\nRequest path parameter, fill the value directly in the request path\nYou can choose to query by external_task_id or task_id\nexternal_task_id\nstring\nOptional\n\nCustomized Task ID for Multi-Image to Video\n\nRequest path parameter, fill the value directly in the request path\nThe external_task_id filled in when creating the task. You can choose to query by external_task_id or task_id\nQuery Task (List)\nGET\n/v1/videos/multi-image2video\ncURL\nCopy\nCollapse\ncurl --request GET \\\n --url 'https://api-singapore.klingai.com/v1/videos/multi-image2video?pageNum=1&pageSize=30' \\\n --header 'Authorization: Bearer '\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error code; Specific definitions can be found in \"Error Code\"\n \"message\": \"string\", // Error message\n \"request_id\": \"string\", // Request ID, generated by the system, used to track requests and troubleshoot problems\n \"data\": [\n {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_status\": \"string\", // Task status, Enum values: submitted, processing, succeed, failed\n \"task_status_msg\": \"string\", // Task status message, displays failure reason when task fails (such as content moderation triggers)\n \"task_info\": { //Task creation parameters\n \"external_task_id\": \"string\" //User-defined task ID\n },\n \"task_result\": {\n \"videos\": [\n {\n \"id\": \"string\", // Generated video ID; globally unique\n \"url\": \"string\", // URL for generating videos (Please note that for security purposes, generated images/videos will be deleted after 30 days. Please save them promptly.)\n \"watermark_url\": \"string\", // Watermarked video download URL, anti-hotlinking format\n \"duration\": \"string\" //Total video duration, unit: s (seconds)\n }\n ]\n },\n \"watermark_info\": {\n \"enabled\": boolean\n },\n \"final_unit_deduction\": \"string\", // Final unit deduction for the task\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit: ms\n \"updated_at\": 1722769557708 //Task update time, Unix timestamp, unit: ms\n }\n ]\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nQuery Parameters\npageNum\nint\nOptional\nDefault to 1\n\nPage number\n\nValue range: [1, 1000]\npageSize\nint\nOptional\nDefault to 30\n\nNumber of items per page\n\nValue range: [1, 500]\nPrevious chapter：Image to Video\nNext chapter：Motion Control\nCreate Task\nQuery Task (Single)\nQuery Task (List)\nThe Kling 3.0 Series Models API is Now Fully Available\n– All in One, One for All！\n\nModels Available in This Release\n\nKling 3.0 Motion Control, Kling Video 3.0, Kling Video 3.0 Omni, Kling Image 3.0, Kling Image 3.0 Omni\n\nRefer to \n\nKey Highlights of the Models\n\n3.0 All-in-One: A unified model for multi-modal input and output.\n\nMost powerful consistency across the universe: Subject consistency (supports cameo, subject with voice control, i2v + subject) and text consistency.\nNarrative control at your fingertips: More freedom, precision, and control—up to 15 seconds long, video scene cuts, ultra-high-definition storyboards/images, custom seconds.\nUpgraded native audio-visual output: Supports multiple speakers and languages (with accents).\n\nKling 3.0 Motion Control\n\nConsistent Facial Identity from any angle\nComplex Emotions faithfully reproduced\nHigh fidelity Restoration, Even with Face Occlusions\nConsistent Facial Clarity Across Dynamic Framing\n\nUser Guide ->\n\nKling Video 3.0\n\nCompared to 2.6, expected improvements:\n\nSupports subject upload in I2V scenarios for enhanced consistency\nSignificant improvement in multi-character referencing, especially for three-person scenarios\nSupports Japanese, Korean, and Spanish in addition to Chinese and English\nCapable of generating certain dialects and accents\nBetter distinction and control over different types of audio (speech, sound effects, BGM)\nImproved text retention in I2V scenarios\nSupports scene transitions, with up to 6 shots and customizable storyboarding\n\nUser Guide ->\n\nKling Video 3.0 Omni\n\nCompared to O1, expected improvements:\n\nNative audio-visual synchronization\nSupports video subject creation\nFurther improved consistency in reference-based tasks, especially for characters and products\nCombined capabilities of reference + storyboarding + audio-visual sync significantly enhance usability\nSupports scene transitions, with up to 6 shots\nExtended generation duration up to 15 seconds\n\nUser Guide ->\n\nKling Image 3.0\n\nHighly consistent feature retention\nPrecise response to detail modifications\nAccurate control over style and tone\nRich imaginative capabilities\n\nUser Guide ->\n\nKling Image 3.0 Omni\n\nEnhanced narrative sense\nNew storyboard image set generation, retaining reference image features with scene relevance\nDirect output of 2K/4K ultra-high-definition images\nFurther improved detail consistency\n\nUser Guide ->\n\nThank you for your support and understanding!\n\nI Got It", "Motion Control": "Motion Control\nMulti-elements to video\nExtend Video\nLip Sync\nAvatar\nText to Audio\nVideo to Audio\nText to Speech\nVoice Clone\nImage Recognize\nElement\nEffects\nEffect Templates\nNEW\nVideo Effects\nImage Generation\nModels\nImage Omni\nImage Generation\nReference to Image\nExtend Image\nAI Multi-Shot\nVirtual Try-On\nOthers\nQuery user info\nPricing\nBilling Info\nPrepaid Resource Packs\nProtocols\nPrivacy Policy of API Service\nTerms of API Service\nAPI Service Level Agreement\nMotion Control\nCreate Task\nPOST\n/v1/videos/motion-control\ncURL\nCopy\nCollapse\ncurl --request POST \\\n --url https://api-singapore.klingai.com/v1/videos/motion-control \\\n --header 'Authorization: Bearer ' \\\n --header 'Content-Type: application/json; charset=utf-8' \\\n --data-raw '{\n \"model_name\": \"kling-v2-6\",\n \"image_url\": \"https://p2-kling.klingai.com/kcdn/cdn-kcdn112452/kling-qa-test/multi-3.ng.png\",\n \"prompt\": \"The girl is wearing a loose gray T-shirt and denim shorts\",\n \"video_url\": \"https://p2-kling.klingai.com/kcdn/cdn-kcdn112452/kling-qa-test/dance.mp4\",\n \"keep_original_sound\": \"yes\",\n \"character_orientation\": \"image\",\n \"mode\": \"pro\",\n \"callback_url\": \"\",\n \"external_task_id\": \"xxx\"\n }'\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes; Specific definitions can be found in \"Error Code\"\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system, used to track requests and troubleshoot problems\n \"data\": {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_info\": { //Task creation parameters\n \"external_task_id\": \"string\" //Customer-defined task ID\n },\n \"task_status\": \"string\", // Task status, Enum values: submitted, processing, succeed, failed\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit: ms\n \"updated_at\": 1722769557708 //Task update time, Unix timestamp, unit: ms\n }\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nRequest Body\nmodel_name\nstring\nOptional\nDefault to kling-v2-6\n\nModel Name\n\nEnum values：\nkling-v2-6\nkling-v3\nprompt\nstring\nOptional\n\nText prompt, can include positive and negative descriptions\n\nCan add elements to the scene, achieve camera movement effects, etc. See Kling \"Motion Control\" User Guide\nCannot exceed 2500 characters\nimage_url\nstring\nRequired\n\nReference image. Characters, background and other elements in generated video will follow this reference.\n\nVideo content requirements:\nCharacter proportions should match the reference motion as much as possible; avoid driving half-body characters with full-body motions\nCharacter should show clear upper body or full body including limbs and head, avoid occlusion\nAvoid extreme orientations (upside down, lying flat, etc.). Character should occupy sufficient screen area\nSupports realistic/stylized characters (including humans/humanoid animals/some pure animals/some humanoid body proportion characters)\nSupports image Base64 encoding or image URL (ensure accessibility)\nImportant: When using Base64, do NOT add any prefix like data:image/png;base64,. Submit only the raw Base64 string.\nCorrect Base64 format:\niVBORw0KGgoAAAANSUhEUgAAAAUA...\nIncorrect Base64 format (with data: prefix):\ndata:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAUA...\nSupported image formats: .jpg / .jpeg / .png\nFile size: ≤10MB, dimensions: 300px ~ 65536px, aspect ratio: 1:2.5 ~ 2.5:1\nvideo_url\nstring\nRequired\n\nURL of reference video. Character actions in generated video will follow this reference.\n\nVideo content requirements:\nCharacter should show clear upper body or full body including all limbs and head, avoid occlusion\nRecommend uploading single-person action video; for 2+ people, actions will be taken from the character with largest screen proportion\nRecommend using real person actions; some stylized characters/humanoid body proportions may work\nVideo should be single continuous shot with character always visible, avoid cuts or camera movements (will be truncated otherwise)\nAvoid overly fast actions; relatively stable actions produce better results\nSupported formats: .mp4 / .mov, file size: ≤100MB, dimensions: 340px ~ 3850px. Validation failures will return error codes.\nDuration limits: minimum 3 seconds, maximum depends on character_orientation:\nWhen character_orientation is \"video\": maximum 30 seconds\nWhen character_orientation is \"image\": maximum 10 seconds\nThe duration range of the uploaded motion reference is from 3 to 30 seconds, in which the generated video length will align with the duration of the uploaded video. If motions are complex or fast-paced, there is a chance that the output may be shorter than the uploaded video duration, as the model can only extract the valid action duration for generation. The minimum extractable continuous action duration is 3 seconds. Please note that in such cases, the consumed credits cannot be refunded. It is recommended to adjust the complexity and speed of the actions accordingly.\nSystem will validate video content and return error codes if issues are found\nelement_list\narray\nOptional\n\nReference Element List based on element ID configuration\n\nLoad with key:value format as follows:\n\"element_list\":[\n { \"element_id\": 829836802793406551 }\n]\nWhen referencing the element, the generated video can only temporarily refer to the orientation of the person in the video.\nCurrently, only one element can be introduced.\n▾\nHide child attributes\nelement_id\nlong\nRequired\n\nElement ID from element library\n\nkeep_original_sound\nstring\nOptional\nDefault to yes\n\nWhether to keep the original sound of the video\n\nEnum values：\nyes\nno\ncharacter_orientation\nstring\nRequired\n\nCharacter orientation in generated video, can match image or video\n\nEnum values：\nimage\nvideo\nimage: Match character orientation in the image; reference video duration must not exceed 10 seconds\nvideo: Match character orientation in the video; reference video duration must not exceed 30 seconds\nWhen referencing the element, the generated video can only temporarily refer to the orientation of the person in the video.\nmode\nstring\nRequired\n\nVideo generation mode\n\nEnum values：\nstd\npro\nstd: Standard Mode - basic mode, cost-effective\npro: Professional Mode (High Quality) - high performance mode, better video quality\n\nSupport varies by model version and video mode. See Capability Map for details.\n\nwatermark_info\nobject\nOptional\n\nWhether to generate watermarked results simultaneously\n\nDefined by the enabled parameter, format:\n \"watermark_info\": { \"enabled\": boolean } \ntrue: generate watermarked result, false: do not generate\nCustom watermarks are not currently supported\ncallback_url\nstring\nOptional\n\nCallback notification URL for task result. If configured, server will notify when task status changes.\n\nFor specific message schema, see Callback Protocol\nexternal_task_id\nstring\nOptional\n\nCustomized Task ID\n\nWill not overwrite system-generated task ID, but supports querying task by this ID\nMust be unique within a single user account\nQuery Task (Single)\nGET\n/v1/videos/motion-control/{id}\ncURL\nCopy\nCollapse\ncurl --request GET \\\n --url https://api-singapore.klingai.com/v1/videos/motion-control/{task_id} \\\n --header 'Authorization: Bearer '\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes; Specific definitions can be found in \"Error Code\"\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system, used to track requests and troubleshoot problems\n \"data\": {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_status\": \"string\", // Task status, Enum values: submitted, processing, succeed, failed\n \"task_status_msg\": \"string\", // Task status information, displaying the failure reason when the task fails (such as triggering the content risk control of the platform, etc.)\n \"task_info\": { //Task creation parameters\n \"external_task_id\": \"string\" //Customer-defined task ID\n },\n \"task_result\": {\n \"videos\": [\n {\n \"id\": \"string\", // Generated video ID; globally unique\n \"url\": \"string\", // URL for generating videos, anti-leech format (To ensure information security, generated images/videos will be cleared after 30 days. Please make sure to save them promptly.)\n \"watermark_url\": \"string\", // Watermarked video download URL, anti-leech format\n \"duration\": \"string\" //Total video duration, unit: s (seconds)\n }\n ]\n },\n \"watermark_info\": {\n \"enabled\": boolean\n \t},\n \"final_unit_deduction\": \"string\", // The deduction units of task\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit: ms\n \"updated_at\": 1722769557708 //Task update time, Unix timestamp, unit: ms\n }\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nPath Parameters\ntask_id\nstring\nOptional\n\nTask ID for Motion Control. Fill the value directly in the request path. Choose either task_id or external_task_id for querying.\n\nexternal_task_id\nstring\nOptional\n\nCustomized Task ID for Motion Control. Fill the value directly in the request path. Choose either task_id or external_task_id for querying.\n\nQuery Task (List)\nGET\n/v1/videos/motion-control\ncURL\nCopy\nCollapse\ncurl --request GET \\\n --url 'https://api-singapore.klingai.com/v1/videos/motion-control?pageNum=1&pageSize=30' \\\n --header 'Authorization: Bearer '\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes; Specific definitions can be found in \"Error Code\"\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system, used to track requests and troubleshoot problems\n \"data\": [\n {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_status\": \"string\", // Task status, Enum values: submitted, processing, succeed, failed\n \"task_status_msg\": \"string\", // Task status information, displaying the failure reason when the task fails (such as triggering the content risk control of the platform, etc.)\n \"task_info\": { //Task creation parameters\n \"external_task_id\": \"string\" //Customer-defined task ID\n },\n \"task_result\": {\n \"videos\": [\n {\n \"id\": \"string\", // Generated video ID; globally unique\n \"url\": \"string\", // URL for generating videos, anti-leech format (To ensure information security, generated images/videos will be cleared after 30 days. Please make sure to save them promptly.)\n \"watermark_url\": \"string\", // Watermarked video download URL, anti-leech format\n \"duration\": \"string\" //Total video duration, unit: s (seconds)\n }\n ]\n },\n \"watermark_info\": {\n \"enabled\": boolean\n },\n \"final_unit_deduction\": \"string\", // The deduction units of task\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit: ms\n \"updated_at\": 1722769557708 //Task update time, Unix timestamp, unit: ms\n }\n ]\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nQuery Parameters\npageNum\nint\nOptional\nDefault to 1\n\nPage number\n\nValue range: [1, 1000]\npageSize\nint\nOptional\nDefault to 30\n\nData volume per page\n\nValue range: [1, 500]\nPrevious chapter：Reference to Video\nNext chapter：Multi-elements to video\nCreate Task\nQuery Task (Single)\nQuery Task (List)\nThe Kling 3.0 Series Models API is Now Fully Available\n– All in One, One for All！\n\nModels Available in This Release\n\nKling 3.0 Motion Control, Kling Video 3.0, Kling Video 3.0 Omni, Kling Image 3.0, Kling Image 3.0 Omni\n\nRefer to \n\nKey Highlights of the Models\n\n3.0 All-in-One: A unified model for multi-modal input a", "Multi-elements to video": "Motion Control\nMulti-elements to video\nExtend Video\nLip Sync\nAvatar\nText to Audio\nVideo to Audio\nText to Speech\nVoice Clone\nImage Recognize\nElement\nEffects\nEffect Templates\nNEW\nVideo Effects\nImage Generation\nModels\nImage Omni\nImage Generation\nReference to Image\nExtend Image\nAI Multi-Shot\nVirtual Try-On\nOthers\nQuery user info\nPricing\nBilling Info\nPrepaid Resource Packs\nProtocols\nPrivacy Policy of API Service\nTerms of API Service\nAPI Service Level Agreement\nMulti-Elements\nInitialize Video for Editing\nPOST\n/v1/videos/multi-elements/init-selection\ncURL\nCopy\nCollapse\ncurl --request POST \\\n --url https://api-singapore.klingai.com/v1/videos/multi-elements/init-selection \\\n --header 'Authorization: Bearer ' \\\n --header 'Content-Type: application/json' \\\n --data '{\n \"video_id\": \"\",\n \"video_url\": \"https://v1-kling.klingai.com/kcdn/cdn-kcdn112452/kling-qa-test/animals-output-5s.mp4\"\n }'\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error code; Specific definitions can be found in \"Error Code\"\n \"message\": \"string\", // Error message\n \"request_id\": \"string\", // Request ID, generated by the system, used for tracking requests and troubleshooting\n \"data\": {\n \"status\": 0, // Rejection code, non-zero indicates recognition failure\n \"session_id\": \"id\", // Session ID, generated during video initialization task, remains unchanged during editing operations, valid for 24 hours\n \"final_unit_deduction\": \"string\", // The deduction units of task\n \"fps\": 30.0, // Frame rate of parsed video, required when fetching selection preview video\n \"original_duration\": 1000, // Duration of parsed video, required when creating task\n \"width\": 720, // Width of parsed video, currently unused\n \"height\": 1280, // Height of parsed video, currently unused\n \"total_frame\": 300, // Total frame count of parsed video, required when creating task\n \"normalized_video\": \"url\" // URL of initialized video\n}\n💡\n\nInitialize the original video before using Multi-elements feature. When replacing or removing elements within the existing video, the relevant elements need to be marked in the video beforehand.\n\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nRequest Body\nvideo_id\nstring\nOptional\n\nThe ID of the video generated by the Kling AI\n\nOnly videos generated within the last 30 days are supported\nOnly supports videos with a duration of ≥2 seconds and ≤5 seconds, or ≥7 seconds and ≤10 seconds\nRelated to the video_url parameter: both video_id and video_url cannot be empty at the same time, and cannot both have values at the same time\nvideo_url\nstring\nOptional\n\nGet link for uploaded video\n\nOnly .mp4/.mov formats are supported\nOnly supports videos with a duration of ≥2 seconds and ≤5 seconds, or ≥7 seconds and ≤10 seconds\nVideo resolution must be between 720px and 2160px (inclusive) in both width and height\nOnly supports videos with frame rates of 24, 30, or 60 fps\nRelated to the video_id parameter: both video_id and video_url cannot be empty at the same time, and cannot both have values at the same time\nAdd Video Selection Area\nPOST\n/v1/videos/multi-elements/add-selection\ncURL\nCopy\nCollapse\ncurl --request POST \\\n --url https://api-singapore.klingai.com/v1/videos/multi-elements/add-selection \\\n --header 'Authorization: Bearer ' \\\n --header 'Content-Type: application/json' \\\n --data '{\n \"session_id\": \"847570360458960960\",\n \"frame_index\": 0,\n \"points\": [\n {\n \"x\": 0.7738498789346246,\n \"y\": 0.297142857142857\n }\n ]\n }'\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error code; Specific definitions can be found in \"Error Code\"\n \"message\": \"string\", // Error message\n \"request_id\": \"string\", // Request ID, generated by the system, used for tracking requests and troubleshooting\n \"data\": {\n \"status\": 0, // Rejection code, non-zero indicates recognition failure\n \"session_id\": \"id\", // Session ID, generated during video initialization task, remains unchanged during editing operations, valid for 24 hours\n \"final_unit_deduction\": \"string\", // The deduction units of task\n \"res\": {\n \"frame_index\": 0,\n \"rle_mask_list\": [{\n \"object_id\": 0,\n \"rle_mask\": {\n \"size\": [720, 1280],\n \"counts\": \"string\"\n },\n \"png_mask\": {\n \"size\": [720, 1280],\n \"base64\": \"string\"\n }\n }]\n }\n }\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nRequest Body\nsession_id\nstring\nRequired\n\nSession ID, generated during the video initialization task and remains unchanged during editing operations\n\nframe_index\nint\nRequired\n\nFrame Number\n\nA maximum of 10 frames can be marked. That is, up to 10 frames can be used to define selection areas in the video\nOnly supports marking 1 frame at a time\npoints\narray\nRequired\n\nClick Coordinates, represented by x and y\n\nValue range: [0, 1], expressed as percentages; [0, 1] represents the top-left corner of the frame\nMultiple points can be marked at once; up to 10 points can be marked on a single frame\n▾\nHide child attributes\nx\nfloat\nRequired\n\nX coordinate [0-1]\n\ny\nfloat\nRequired\n\nY coordinate [0-1]\n\nSample Code\nDecoding Image Segmentation Result\nexport type RLEObject = {\n size: [h: number, w: number]\n counts: string\n}\ntype RLE = {\n h: number\n w: number\n m: number\n binaries: number[]\n}\nexport function decode(rleObj: RLEObject) {\n const [h, w] = rleObj.size\n const R: RLE = { h, w, m: 0, binaries: [0] }\n rleFrString(R, rleObj.counts)\n const unitArray = new Uint8Array(h * w)\n rleDecode(R, unitArray)\n return unitArray\n}\nfunction rleDecode(R: RLE, M: Uint8Array) {\n let j\n let k\n let p = 0\n let v = false\n for (j = 0; j < R.m; j++) {\n for (k = 0; k < R.binaries[j]; k++) {\n const x = Math.floor(p / R.h)\n const y = p % R.h\n M[y * R.w + x] = v === false ? 0 : 1 // Note: y * width + x indicates row-major (horizontal) layout.\n p++\n }\n v = !v\n }\n}\nfunction rleFrString(R: RLE, s: string) {\n let m = 0\n let p = 0\n let k\n let x\n let more\n const binaries = []\n while (s[p]) {\n x = 0\n k = 0\n more = 1\n while (more) {\n const c = s.charCodeAt(p) - 48\n x |= (c & 0x1f) << (5 * k)\n more = c & 0x20\n p++\n k++\n if (!more && c & 0x10) {\n x |= -1 << (5 * k)\n }\n }\n if (m > 2) {\n x += binaries[m - 2]\n }\n binaries[m++] = x\n }\n R.m = m\n R.binaries = binaries\n}\nRendering the Segmentation Mask Layer\n// height refers to the video height and width refers to the video width\nfunction drawMask(rleMask: string, height: number, width: number) {\n if (!canvasRef.value) return\n const ctx = canvasRef.value.getContext('2d')\n if (!ctx) return\n\n const decodeData = decode({ counts: rleMask, size: [height, width] })\n const imageData = ctx.createImageData(width, height)\n for (let y = 0; y < height; y++) {\n for (let x = 0; x < width; x++) {\n const index = y * width + x\n if (decodeData[index]) {\n const imageIndex = index * 4\n // Set pixel color: red, green, blue, alpha\n imageData.data[imageIndex] = 116 // red\n imageData.data[imageIndex + 1] = 255 // green\n imageData.data[imageIndex + 2] = 82 // blue\n imageData.data[imageIndex + 3] = 163 // alpha\n }\n }\n }\n ctx.putImageData(imageData, 0, 0)\n}\n\nDelete Video Selection Area\nPOST\n/v1/videos/multi-elements/delete-selection\ncURL\nCopy\nCollapse\ncurl --request POST \\\n --url https://api-singapore.klingai.com/v1/videos/multi-elements/delete-selection \\\n --header 'Authorization: Bearer ' \\\n --header 'Content-Type: application/json' \\\n --data '{\n \"session_id\": \"847570360458960960\",\n \"frame_index\": 0,\n \"points\": [\n {\n \"x\": 0.7738498789346246,\n \"y\": 0.297142857142857\n }\n ]\n }'\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error code; Specific definitions can be found in \"Error Code\"\n \"message\": \"string\", // Error message\n \"request_id\": \"string\", // Request ID, generated by the system, used for tracking requests and troubleshooting\n \"data\": {\n \"status\": 0, // Rejection code, non-zero indicates recognition failure\n \"session_id\": \"id\", // Session ID, generated during video initialization task, remains unchanged during editing operations, valid for 24 hours\n \"final_unit_deduction\": \"string\", // The deduction units of task\n \"res\": {\n \"frame_index\": 0,\n \"rle_mask_list\": [{\n \"object_id\": 0,\n \"rle_mask\": {\n \"size\": [720, 1280],\n \"counts\": \"string\"\n },\n \"png_mask\": {\n \"size\": [720, 1280],\n \"base64\": \"string\"\n }\n }]\n }\n }\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nRequest Body\nsession_id\nstring\nRequired\n\nSession ID, generated during the video initialization task and remains unchanged during editing operations\n\nframe_index\nint\nRequired\n\nFrame Number\n\npoints\narray\nRequired\n\nClick Coordinates to delete, represented by x and y\n\nValue range: [0, 1], expressed as percentages; [0, 1] represents the top-left corner of the frame\nMultiple points can be provided at once\nCoordinates must exactly match those used when adding the video selection area\n▾\nHide child attributes\nx\nfloat\nRequired\n\nX coordinate [0-1]\n\ny\nfloat\nRequired\n\nY coordinate [0-1]\n\nClear Video Selection\nPOST\n/v1/videos/multi-elements/clear-selection\ncURL\nCopy\nCollapse\ncurl --request POST \\\n --url https://api-singapore.klingai.com/v1/videos/multi-elements/clear-selection \\\n --header 'Authorization: Bearer ' \\\n --header 'Content-Type: application/json' \\\n --data '{\n \"session_id\": \"847570360458960960\"\n }'\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error code; Specific definitions can be found in \"Error Code\"\n \"message\": \"string\", // Error message\n \"request_id\": \"string\", // Request ID, generated by the system, used for tracking requests and troubleshooting\n \"data\": {\n \"status\": 0, // Rejection code, non-zero indicates recognition failure\n \"session_id\": \"id\" // Session ID, generated during video initialization task, remains unchanged during editing operations, valid for 24 hours\n \"final_unit_deduction\": \"string\", // The deduction units of task\n }\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nRequest Body\nsession_id\nstring\nRequired\n\nSession ID, generated during the video initialization task and remains unchanged during editing operations\n\nPreview Video with Selected Areas\nPOST\n/v1/videos/multi-elements/preview-selection\ncURL\nCopy\nCollapse\ncurl --request POST \\\n --url https://api-singapore.klingai.com/v1/videos/multi-elements/preview-selection \\\n --header 'Authorization: Bearer ' \\\n --header 'Content-Type: application/json' \\\n --data '{\n \"session_id\": \"847570360458960960\"\n }'\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error code; Specific definitions can be found in \"Error Code\"\n \"message\": \"string\", // Error message\n \"request_id\": \"string\", // Request ID, generated by the system, used for tracking requests and troubleshooting\n \"data\": {\n \"status\": 0, // Rejection code, non-zero indicates recognition failure\n \"session_id\": \"id\", // Session ID, generated during video initialization task, remains unchanged during editing operations, valid for 24 hours\n \"final_unit_deduction\": \"string\", // The deduction units of task\n \"res\": {\n \"video\": \"url\", // Video with mask\n \"video_cover\": \"url\", // Cover image of video with mask\n \"tracking_output\": \"url\" // Mask result for each frame in image segmentation results\n }\n }\n}\nRequest Header\nConte", "Extend Video": "Video Extension\nCreate Task\nPOST\n/v1/videos/video-extend\ncURL\nCopy\nCollapse\ncurl --request POST \\\n --url https://api-singapore.klingai.com/v1/videos/video-extend \\\n --header 'Authorization: Bearer ' \\\n --header 'Content-Type: application/json' \\\n --data '{\n \"prompt\": \"A puppy appears\",\n \"video_id\": \"743211632612511839\",\n \"negative_prompt\": \"\",\n \"callback_url\": \"\"\n }'\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes; Specific definitions can be found in Error codes\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system, is used to track requests and troubleshoot problems\n \"data\": {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_status\": \"string\", // Task status, Enum values: submitted, processing, succeed, failed\n \"task_info\":{ // Task creation parameters\n \"external_task_id\": \"string\" // Customer-defined task ID\n },\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit: ms\n \"updated_at\": 1722769557708 //Task update time, Unix timestamp, unit: ms\n }\n}\nNote 1: Video extension refers to extending the duration of text-to-video/image-to-video results. Each extension can add 4 to 5 seconds, and the model and mode used cannot be selected; they must be the same as the source video.\nNote 2: Videos that have been extended can be extended again, but the total video duration cannot exceed 3 minutes.\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nRequest Body\nvideo_id\nstring\nRequired\n\nVideo ID\n\nSupports video IDs generated by the text-to-video, the image-to-video and the video extension interface (it cannot exceed 3 minutes).\nOnly videos generated by V1.0, V1.5, and V1.6 models are supported.\n\nPlease note that based on the current cleanup policy, videos will be cleared 30 days after generation, and extension will not be possible.\n\nprompt\nstring\nOptional\n\nText Prompt\n\nCannot exceed 2500 characters\n\nnegative_prompt\nstring\nOptional\n\nNegative text prompt\n\nCannot exceed 2500 characters\n\ncfg_scale\nfloat\nOptional\nDefault to 0.5\n\nPrompt reference strength. The higher the value, the stronger the reference to the prompt.\n\nValue range: [0, 1]\n\nwatermark_info\nobject\nOptional\n\nWhether to generate watermarked results simultaneously\n\nDefined by the enabled parameter, format:\n \"watermark_info\": { \"enabled\": boolean } \ntrue: generate watermarked result, false: do not generate\nCustom watermarks are not currently supported\ncallback_url\nstring\nOptional\n\nThe callback notification address for the task results. If configured, the server will actively notify when the task status changes.\n\nFor specific message schema, see Callback Protocol\nexternal_task_id\nstring\nOptional\n\nCustomized Task ID\n\nUser-defined task ID. It will not override the system-generated task ID, but supports querying tasks by this ID\nPlease note that it must be unique for each user\nQuery Task (Single)\nGET\n/v1/videos/video-extend/{id}\ncURL\nCopy\nCollapse\ncurl --request GET \\\n --url https://api-singapore.klingai.com/v1/videos/video-extend/{task_id} \\\n --header 'Authorization: Bearer '\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes; Specific definitions can be found in Error codes\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system, is used to track requests and troubleshoot problems\n \"data\": {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_status\": \"string\", // Task status, Enum values: submitted, processing, succeed, failed\n \"task_status_msg\": \"string\", // Task status information, displaying the failure reason when the task fails (such as triggering the content risk control of the platform, etc.)\n \"task_info\": {\n \"parent_video\": { //Parameters information when the task is created\n \"id\": \"string\", //Video ID before the extension；globally unique\n \"url\": \"string\", //URL for generating images(To ensure information security, generated images/videos will be cleared after 30 days. Please make sure to save them promptly.)\n \"duration\": \"string\" //Original video duration, unit: s (seconds)\n },\n \"external_task_id\": \"string\" // Customer-defined task ID\n },\n \"task_result\": {\n \"videos\": [\n {\n \"id\": \"string\", // Generated video ID; globally unique, will be cleared after 30 days\n \"url\": \"string\", // URL for generating videos (To ensure information security, generated images/videos will be cleared after 30 days. Please make sure to save them promptly.)\n \"watermark_url\": \"string\", // Watermarked video download URL, anti-leech format\n \"duration\": \"string\" //Total video duration, unit: s (seconds)\n }\n ]\n },\n \"watermark_info\": {\n \"enabled\": boolean\n },\n \"final_unit_deduction\": \"string\", // The deduction units of task\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit: ms\n \"updated_at\": 1722769557708 //Task update time, Unix timestamp, unit: ms\n }\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nPath Parameters\ntask_id\nstring\nOptional\n\nTask ID for Video Generation. Fill the value directly in the request path. Choose either task_id or external_task_id for querying.\n\nexternal_task_id\nstring\nOptional\n\nCustomized Task ID for Video Generation. Fill the value directly in the request path. Choose either task_id or external_task_id for querying.\n\nQuery Task (List)\nGET\n/v1/videos/video-extend\ncURL\nCopy\nCollapse\ncurl --request GET \\\n --url 'https://api-singapore.klingai.com/v1/videos/video-extend?pageNum=1&pageSize=30' \\\n --header 'Authorization: Bearer '\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes; Specific definitions can be found in Error codes\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system, is used to track requests and troubleshoot problems\n \"data\": [\n {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_status\": \"string\", // Task status, Enum values: submitted, processing, succeed, failed\n \"task_status_msg\": \"string\", // Task status information, displaying the failure reason when the task fails (such as triggering the content risk control of the platform, etc.)\n \"task_info\": {\n \"parent_video\": { //Parameters information when the task is created\n \"id\": \"string\", //Video ID before the extension；globally unique\n \"url\": \"string\", //URL for generating images(To ensure information security, generated images/videos will be cleared after 30 days. Please make sure to save them promptly.)\n \"duration\": \"string\" //Original video duration, unit: s (seconds)\n },\n \"external_task_id\": \"string\" // Customer-defined task ID\n },\n \"task_result\": {\n \"videos\": [\n {\n \"id\": \"string\", // Generated video ID; globally unique, will be cleared after 30 days\n \"url\": \"string\", // URL for generating videos (To ensure information security, generated images/videos will be cleared after 30 days. Please make sure to save them promptly.)\n \"watermark_url\": \"string\", // Watermarked video download URL, anti-leech format\n \"duration\": \"string\" //Total video duration, unit: s (seconds)\n }\n ]\n },\n \"watermark_info\": {\n \"enabled\": boolean\n },\n \"final_unit_deduction\": \"string\", // The deduction units of task\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit: ms\n \"updated_at\": 1722769557708 //Task update time, Unix timestamp, unit: ms\n }\n ]\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nQuery Parameters\npageNum\nint\nOptional\nDefault to 1\n\nPage number\n\nValue range: [1, 1000]\n\npageSize\nint\nOptional\nDefault to 30\n\nNumber of items per page\n\nValue range: [1, 500]\n\nPrevious chapter：Multi-elements to video\nNext chapter：Lip Sync\nCreate Task\nQuery Task (Single)\nQuery Task (List)\nThe Kling 3.0 Series Models API is Now Fully Available\n– All in One, One for All！\n\nModels Available in This Release\n\nKling 3.0 Motion Control, Kling Video 3.0, Kling Video 3.0 Omni, Kling Image 3.0, Kling Image 3.0 Omni\n\nRefer to \n\nKey Highlights of the Models\n\n3.0 All-in-One: A unified model for multi-modal input and output.\n\nMost powerful consistency across the universe: Subject consistency (supports cameo, subject with voice control, i2v + subject) and text consistency.\nNarrative control at your fingertips: More freedom, precision, and control—up to 15 seconds long, video scene cuts, ultra-high-definition storyboards/images, custom seconds.\nUpgraded native audio-visual output: Supports multiple speakers and languages (with accents).\n\nKling 3.0 Motion Control\n\nConsistent Facial Identity from any angle\nComplex Emotions faithfully reproduced\nHigh fidelity Restoration, Even with Face Occlusions\nConsistent Facial Clarity Across Dynamic Framing\n\nUser Guide ->\n\nKling Video 3.0\n\nCompared to 2.6, expected improvements:\n\nSupports subject upload in I2V scenarios for enhanced consistency\nSignificant improvement in multi-character referencing, especially for three-person scenarios\nSupports Japanese, Korean, and Spanish in addition to Chinese and English\nCapable of generating certain dialects and accents\nBetter distinction and control over different types of audio (speech, sound effects, BGM)\nImproved text retention in I2V scenarios\nSupports scene transitions, with up to 6 shots and customizable storyboarding\n\nUser Guide ->\n\nKling Video 3.0 Omni\n\nCompared to O1, expected improvements:\n\nNative audio-visual synchronization\nSupports video subject creation\nFurther improved consistency in reference-based tasks, especially for characters and products\nCombined capabilities of reference + storyboarding + audio-visual sync significantly enhance usability\nSupports scene transitions, with up to 6 shots\nExtended generation duration up to 15 seconds\n\nUser Guide ->\n\nKling Image 3.0\n\nHighly consistent feature retention\nPrecise response to detail modifications\nAccurate control over style and tone\nRich imaginative capabilities\n\nUser Guide ->\n\nKling Image 3.0 Omni\n\nEnhanced narrative sense\nNew storyboard image set generation, retaining reference image features with scene relevance\nDirect output of 2K/4K ultra-high-definition images\nFurther improved detail consistency\n\nUser Guide ->\n\nThank you for your support and understanding!\n\nI Got It", "Lip Sync": "Motion Control\nMulti-elements to video\nExtend Video\nLip Sync\nAvatar\nText to Audio\nVideo to Audio\nText to Speech\nVoice Clone\nImage Recognize\nElement\nEffects\nEffect Templates\nNEW\nVideo Effects\nImage Generation\nModels\nImage Omni\nImage Generation\nReference to Image\nExtend Image\nAI Multi-Shot\nVirtual Try-On\nOthers\nQuery user info\nPricing\nBilling Info\nPrepaid Resource Packs\nProtocols\nPrivacy Policy of API Service\nTerms of API Service\nAPI Service Level Agreement\nLip-Sync\nIdentify Face\nPOST\n/v1/videos/identify-face\ncURL\nCopy\nCollapse\ncurl --request POST \\\n --url https://api-singapore.klingai.com/v1/videos/identify-face \\\n --header 'Authorization: Bearer ' \\\n --header 'Content-Type: application/json' \\\n --data '{\n \"video_url\": \"https://p1-kling.klingai.com/kcdn/cdn-kcdn112452/kling-qa-test/kling20260206mp4.mp4\",\n \"video_id\": \"\"\n }'\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes; Specific definitions can be found in \"Error Code\"\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system, used to track requests and troubleshoot problems\n \"data\": {\n \"session_id\": \"id\", // Session ID\n \"final_unit_deduction\": \"string\", // The deduction units of task\n \"face_data\": [ //Face data list\n {\n \"face_id\": \"string\", // Face ID\n \"face_image\": \"url\", // Face image URL\n \"start_time\": 0, // Face appearance start time, unit: ms\n \"end_time\": 5200 //Face appearance end time, unit: ms\n }\n ]\n }\n}\n\nIdentify faces in the video for lip-sync processing.\n\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nRequest Body\nvideo_id\nstring\nOptional\n\nThe ID of the video generated by Kling AI\n\nUsed to specify the video and determine whether it can be used for lip-sync services.\nThis parameter and 'video_url' are mutually exclusive—only one can be filled, and neither can be left empty.\nOnly supports videos generated within the last 30 days with a duration of no more than 60 seconds.\nvideo_url\nstring\nOptional\n\nThe URL of the video\n\nUsed to specify the video and determine whether it can be used for lip-sync services.\nThis parameter and 'video_id' are mutually exclusive—only one can be filled, and neither can be left empty.\nSupported video formats: .mp4/.mov, file size ≤100MB, duration 2s–60s, resolution 720p or 1080p, with both width and height between 512px–2160px. If validation fails, an error code will be returned.\nThe system checks video content—if issues are detected, an error code will be returned.\nCreate Task\nPOST\n/v1/videos/advanced-lip-sync\ncURL\nCopy\nCollapse\ncurl --request POST \\\n --url https://api-singapore.klingai.com/v1/videos/advanced-lip-sync \\\n --header 'Authorization: Bearer ' \\\n --header 'Content-Type: application/json' \\\n --data '{\n \"session_id\": \"850508686686064678\",\n \"face_choose\": [\n {\n \"face_id\": \"0\",\n \"sound_file\": \"https://p1-kling.klingai.com/kcdn/cdn-kcdn112452/kling-qa-test/go-to-world.mp3\",\n \"sound_insert_time\": 1000,\n \"sound_start_time\": 0,\n \"sound_end_time\": 3000,\n \"sound_volume\": 2,\n \"original_audio_volume\": 2\n }\n ],\n \"external_task_id\": \"\",\n \"callback_url\": \"\"\n }'\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes; Specific definitions can be found in \"Error Code\"\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system, used to track requests and troubleshoot problems\n \"data\": {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_info\": { //Task creation parameters\n \"external_task_id\": \"string\" //User-defined task ID\n },\n \"task_status\": \"string\", // Task status, Enum values: submitted, processing, succeed, failed\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit: ms\n \"updated_at\": 1722769557708 //Task update time, Unix timestamp, unit: ms\n }\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nRequest Body\nsession_id\nstring\nRequired\n\nSession ID generated during the identify face API. It remains unchanged during the selection/editing process.\n\nface_choose\narray\nRequired\n\nSpecified Face for Lip-Sync\n\nIncludes Face ID, lip movement reference data, etc.\nCurrently only supports one person lip-sync.\n▾\nHide child attributes\nface_id\nstring\nRequired\n\nFace ID\n\nReturned by the facial recognition interface.\naudio_id\nstring\nOptional\n\nSound ID Generated via TTS API\n\nOnly supports audio generated within the last 30 days with a duration of no less than 2 seconds and no more than 60 seconds.\nEither audio_id or sound_file must be provided (mutually exclusive; cannot be empty or both populated).\nsound_file\nstring\nOptional\n\nSound File\n\nSupports Base64-encoded audio or accessible audio URL.\nAccepted formats: .mp3/.wav/.m4a/.aac (max 5MB). Format mismatches or oversized files will return error codes.\nOnly supports audio with a duration of no less than 2 seconds and no more than 60 seconds.\nEither audio_id or sound_file must be provided (mutually exclusive; cannot be empty or both populated).\nThe system will verify the audio content and return error codes if there are any problems.\nsound_start_time\nlong\nRequired\n\nTime point to start cropping sound\n\nBased on the original sound start time, the start time is 0'0\", units: ms\nThe sound before the starting point will be cropped, and the cropped sound must not be shorter than 2 seconds.\nsound_end_time\nlong\nRequired\n\nTime point to stop cropping sound\n\nBased on the original sound start time, the start time is 0'0\", units: ms\nThe sound after the end point will be cropped, and the cropped sound must not be shorter than 2 seconds.\nThe end point time shouldn't be later than the total duration of the original sound.\nsound_insert_time\nlong\nRequired\n\nThe time for inserting cropped sound\n\nBased on the original video start time, the start time is 0'0\", units: ms\nThe time range for inserting sound should overlap with the face's lip-sync time interval for at least 2 seconds.\nThe start time for inserting sound must not be earlier than the start time of the video, and the end time for inserting sound must not be later than the end time of the video.\nsound_volume\nfloat\nOptional\nDefault to 1\n\nVolume Controls (higher values = louder)\n\nValue range: [0, 2]\noriginal_audio_volume\nfloat\nOptional\nDefault to 1\n\nOriginal video volume (higher values = louder)\n\nValue range: [0, 2]\nNo effect if source video is silent.\nwatermark_info\nobject\nOptional\n\nWhether to generate watermarked results simultaneously\n\nDefined by the enabled parameter, format:\n \"watermark_info\": { \"enabled\": boolean } \ntrue: generate watermarked result, false: do not generate\nCustom watermarks are not currently supported\nexternal_task_id\nstring\nOptional\n\nCustom Task ID\n\nUser-defined task ID. It will not override the system-generated task ID, but supports querying tasks by this ID.\nPlease note that uniqueness must be ensured for each user.\ncallback_url\nstring\nOptional\n\nThe callback notification address for the result of this task. If configured, the server will actively notify when the task status changes.\n\nFor specific message schema, see Callback Protocol\nQuery Task (Single)\nGET\n/v1/videos/advanced-lip-sync/{id}\ncURL\nCopy\nCollapse\ncurl --request GET \\\n --url https://api-singapore.klingai.com/v1/videos/advanced-lip-sync/{task_id} \\\n --header 'Authorization: Bearer '\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes; Specific definitions can be found in \"Error Code\"\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system, used to track requests and troubleshoot problems\n \"data\": {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_status\": \"string\", // Task status, Enum values: submitted, processing, succeed, failed\n \"task_status_msg\": \"string\", // Task status message, displaying the failure reason when the task fails (such as triggering the content risk control of the platform, etc.)\n \"task_info\": { //Task creation parameters\n \"parent_video\": { //Original video information\n \"id\": \"string\", // Original video ID\n \"url\": \"string\", // Original video URL\n \"duration\": \"string\" //Original video duration, unit: s\n }\n },\n \"task_result\": { //Task result\n \"videos\": [ //Generated video list\n {\n \"id\": \"string\", // Generated video ID; globally unique\n \"url\": \"string\", // URL for generating videos (Please note that for security purposes, generated images/videos will be deleted after 30 days. Please save them promptly.)\n \"watermark_url\": \"string\", // Watermarked video download URL, anti-hotlinking format\n \"duration\": \"string\" //Total video duration, unit: s\n }\n ]\n },\n \"watermark_info\": {\n \"enabled\": boolean //Whether watermark is enabled\n },\n \"final_unit_deduction\": \"string\", // The deduction units of task\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit: ms\n \"updated_at\": 1722769557708 //Task update time, Unix timestamp, unit: ms\n }\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nPath Parameters\ntask_id\nstring\nOptional\n\nTask ID for Video Generation - Lip-Sync. Fill the value directly in the request path.\n\nQuery Task (List)\nGET\n/v1/videos/advanced-lip-sync\ncURL\nCopy\nCollapse\ncurl --request GET \\\n --url 'https://api-singapore.klingai.com/v1/videos/advanced-lip-sync?pageNum=1&pageSize=30' \\\n --header 'Authorization: Bearer '\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes; Specific definitions can be found in \"Error Code\"\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system, used to track requests and troubleshoot problems\n \"data\": [\n {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_status\": \"string\", // Task status, Enum values: submitted, processing, succeed, failed\n \"task_status_msg\": \"string\", // Task status message, displaying the failure reason when the task fails (such as triggering the content risk control of the platform, etc.)\n \"task_info\": { //Task creation parameters\n \"parent_video\": { //Original video information\n \"id\": \"string\", // Original video ID\n \"url\": \"string\", // Original video URL\n \"duration\": \"string\" //Original video duration, unit: s\n }\n },\n \"task_result\": { //Task result\n \"videos\": [ //Generated video list\n {\n \"id\": \"string\", // Generated video ID; globally unique\n \"url\": \"string\", // URL for generating videos (Please note that for security purposes, generated images/videos will be deleted after 30 days. Please save them promptly.)\n \"watermark_url\": \"string\", // Watermarked video download URL, anti-hotlinking format\n \"duration\": \"string\" //Total video duration, unit: s\n }\n ]\n },\n \"watermark_info\": {\n \"enabled\": boolean //Whether watermark is enabled\n },\n \"final_unit_deduction\": \"string\", // The deduction units of task\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit: ms\n \"updated_at\": 1722769557708 //Task update time, Unix timestamp, unit: ms\n }\n ]\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nQuery Parameters\npageNum\nint\nOptional\nDefault to 1\n\nPage number\n\nValue range: [1, 1000]\npageSize\nint\nOptional\nDefault to 30\n\nNumber of items per page\n\nValue ran", "Image Omni": "Motion Control\nMulti-elements to video\nExtend Video\nLip Sync\nAvatar\nText to Audio\nVideo to Audio\nText to Speech\nVoice Clone\nImage Recognize\nElement\nEffects\nEffect Templates\nNEW\nVideo Effects\nImage Generation\nModels\nImage Omni\nImage Generation\nReference to Image\nExtend Image\nAI Multi-Shot\nVirtual Try-On\nOthers\nQuery user info\nPricing\nBilling Info\nPrepaid Resource Packs\nProtocols\nPrivacy Policy of API Service\nTerms of API Service\nAPI Service Level Agreement\nOmni-Image\nCreate Task\nPOST\n/v1/images/omni-image\ncURL\nCopy\nCollapse\ncurl --request POST \\\n --url https://api-singapore.klingai.com/v1/images/omni-image \\\n --header 'Authorization: Bearer ' \\\n --header 'Content-Type: application/json' \\\n --data '{\n \"model_name\": \"kling-image-o1\",\n \"prompt\": \"Merge all the people in the images into the <<>> image\",\n \"element_list\": [\n {\n \"element_id\": 829836802793406551\n }\n ],\n \"image_list\": [\n {\n \"image\": \"https://v1-kling.klingai.com/kcdn/cdn-kcdn112452/kling-qa-test/multi-4.png\"\n },\n {\n \"image\": \"https://p2-kling.klingai.com/kcdn/cdn-kcdn112452/kling-qa-test/video_effects/1.png\"\n },\n {\n \"image\": \"https://p2-kling.klingai.com/kcdn/cdn-kcdn112452/kling-qa-test/video_effects/4.png\"\n }\n ],\n \"resolution\": \"2k\",\n \"n\": 1,\n \"aspect_ratio\": \"3:2\"\n }'\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error code; Specific definitions can be found in Error codes\n \"message\": \"string\", // Error message\n \"request_id\": \"string\", // Request ID, system-generated, used for tracking requests and troubleshooting\n \"data\": {\n \"task_id\": \"string\", // Task ID, system-generated\n \"task_info\": { //Task creation parameters\n \"external_task_id\": \"string\" //User-defined task ID\n },\n \"task_status\": \"string\", // Task status, Enum values: submitted, processing, succeed, failed\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit: ms\n \"updated_at\": 1722769557708 //Task update time, Unix timestamp, unit: ms\n }\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nRequest Body\nmodel_name\nstring\nOptional\nDefault to kling-image-o1\n\nModel Name\n\nEnum values：\nkling-image-o1\nkling-v3-omni\nprompt\nstring\nRequired\n\nText prompt words, which can include positive and negative descriptions\n\nThe prompt words can be templated to meet different image generation needs\nMust not exceed 2,500 characters\nSpecify an image in the format of <<<>>>, such as <<>>\nThe scope of abilities can be found in the user manual: KLING Omni Model User Guide\nimage_list\narray\nOptional\n\nReference Image List\n\nLoad with key:value format as follows:\n\"image_list\":[\n { \"image\":\"image_url\" }\n]\nSupports inputting image Base64 encoding or image URL (ensure accessibility)\nSupported image formats: .jpg / .jpeg / .png\nImage file size must not exceed 10MB. Image dimensions must be at least 300px. Aspect ratio must be between 1:2.5 and 2.5:1\nThe sum of reference elements and reference images must not exceed 10\nThe value of image_url parameter must not be empty\n▾\nHide child attributes\nimage\nstring\nRequired\n\nImage URL or Base64 string\n\nelement_list\narray\nOptional\n\nReference Element List based on element ID configuration\n\nLoad with key:value format as follows:\n\"element_list\":[\n { \"element_id\": 829836802793406551 }\n]\nThe sum of reference elements and reference images must not exceed 10\n\nThe support range for different model versionsvaries. For details, see Capability Map\n\n▾\nHide child attributes\nelement_id\nlong\nRequired\n\nElement ID from element library\n\nresolution\nstring\nOptional\nDefault to 1k\n\nImage generation resolution\n\nEnum values：\n1k\n2k\n4k\n1k: 1K standard definition\n2k: 2K high-res\n4k: 4K high-res\n\nThe support range for different model versionsvaries. For details, see Capability Map\n\nresult_type\nstring\nOptional\nDefault to single\n\nControl whether to generate a single image or a series of images\n\nEnum values：\nsingle\nseries\n\nThe support range for different model versionsvaries. For details, see Capability Map\n\nn\nint\nOptional\nDefault to 1\n\nNumber of generated images\n\nValue range: [1, 9]\nWhen result_type is series, this parameter is invalid\nseries_amount\nint\nOptional\nDefault to 4\n\nNumber of images in a series\n\nValue range: [2, 9]\nWhen result_type is single, this parameter is invalid\n\nThe support range for different model versionsvaries. For details, see Capability Map\n\naspect_ratio\nstring\nOptional\nDefault to auto\n\nAspect ratio of the generated images (width:height)\n\nEnum values：\n16:9\n9:16\n1:1\n4:3\n3:4\n3:2\n2:3\n21:9\nauto\nauto is to intelligently aspect ratio of the generated image based on incoming content.\nWhen generating a new image based on the aspect ratio of the original image, this parameter is invalid\n\nThe support range for different model versionsvaries. For details, see Capability Map\n\nwatermark_info\nobject\nOptional\n\nWhether to generate watermarked results simultaneously\n\nDefined by the enabled parameter, format:\n \"watermark_info\": { \"enabled\": boolean } \ntrue: generate watermarked result, false: do not generate\nCustom watermarks are not currently supported\ncallback_url\nstring\nOptional\n\nThe callback notification address for the result of this task. If configured, the server will actively notify when the task status changes\n\nFor the specific message schema, see Callback Protocol\nexternal_task_id\nstring\nOptional\n\nCustomized Task ID\n\nUser-defined task ID. It will not override the system-generated task ID, but supports querying tasks by this ID\nPlease note that it must be unique for each user\nInvocation examples\nImage generation with element\ncurl --location 'https://xxx/v1/images/generations' \\\n--header 'Authorization: Bearer xxx' \\\n--header 'Content-Type: application/json' \\\n--data '{\n \"model_name\": \"kling-v3-omni\",\n \"prompt\": \"Generate a recommended cover for each subject <> based on the style of the reference image <>\",\n \"element_list\": [\n {\n \"element_id\": \"160\"\n },\n {\n \"element_id\": \"161\"\n }\n ],\n \"image_list\": [\n {\n \"image\": \"xxx\"\n },\n {\n \"image\": \"xxx\"\n }\n ],\n \"resolution\": \"2k\",\n \"result_type\": \"series\",\n \"series_amount\": 2,\n \"aspect_ratio\": \"auto\",\n \"external_task_id\": \"\",\n \"callback_url\": \"\"\n }'\nQuery Task (Single)\nGET\n/v1/images/omni-image/{id}\ncURL\nCopy\nCollapse\ncurl --request GET \\\n --url https://api-singapore.klingai.com/v1/images/omni-image/{id} \\\n --header 'Authorization: Bearer ' \\\n --header 'Content-Type: application/json'\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes; specific definitions see Error codes\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system, for tracking and troubleshooting\n \"data\": {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_status\": \"string\", // Task status: submitted, processing, succeed, failed\n \"task_status_msg\": \"string\", // Task status message, failure reason when task fails\n \"task_info\": { //Task creation parameters\n \"external_task_id\": \"string\" //Customer-defined task ID\n },\n \"task_result\": {\n \"result_type\": \"single\",\n \"images\": [\n {\n \"index\": 0, // Image number\n \"url\": \"string\", // URL for generated image, anti-hotlinking format (Generated images/videos will be cleared after 30 days. Please save promptly.)\n \"watermark_url\": \"string\" // Watermarked image download URL, anti-hotlinking format\n }\n ],\n \"series_images\": [\n {\n \"index\": 0, // Series-image sequence number\n \"url\": \"string\", // URL for generated image, anti-hotlinking format\n \"watermark_url\": \"string\" // Watermarked image download URL\n }\n ]\n },\n \"watermark_info\": { \"enabled\": boolean },\n \"final_unit_deduction\": \"string\", // Final unit deduction for the task\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit: ms\n \"updated_at\": 1722769557708 //Task update time, Unix timestamp, unit: ms\n }\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nPath Parameters\ntask_id\nstring\nRequired\n\nThe task ID for image generation. Request path parameter, directly fill the value in the request path. Query by external_task_id or task_id.\n\nexternal_task_id\nstring\nOptional\n\nCustomized Task ID for audio generation\n\nThe external_task_id filled in when creating the task. You can choose to query by external_task_id or task_id\nWhen creating a task, you can choose to query by external_task_id or task_id.\nQuery Task (List)\nGET\n/v1/images/omni-image\ncURL\nCopy\nCollapse\ncurl --request GET \\\n --url 'https://api-singapore.klingai.com/v1/images/omni-image?pageNum=1&pageSize=30' \\\n --header 'Authorization: Bearer ' \\\n --header 'Content-Type: application/json'\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes; specific definitions see Error codes\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system\n \"data\": [\n {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_status\": \"string\", // Task status: submitted, processing, succeed, failed\n \"task_status_msg\": \"string\", // Task status message\n \"task_info\": { //Task creation parameters\n \"external_task_id\": \"string\" //Customer-defined task ID\n },\n \"final_unit_deduction\": \"string\", // Final unit deduction for the task\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit: ms\n \"updated_at\": 1722769557708, // Task update time, Unix timestamp, unit: ms\n \"watermark_info\": { \"enabled\": boolean },\n \"task_result\": {\n \"result_type\": \"single\",\n \"images\": [\n {\n \"index\": 0, // Image number\n \"url\": \"string\", // URL for generated image, anti-hotlinking format (Generated images/videos will be cleared after 30 days. Please save promptly.)\n \"watermark_url\": \"string\" // Watermarked image download URL\n }\n ],\n \"series_images\": [\n {\n \"index\": 0, // Series-image sequence number\n \"url\": \"string\", // URL for generated image, anti-hotlinking format\n \"watermark_url\": \"string\" // Watermarked image download URL\n }\n ]\n }\n }\n ]\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nQuery Parameters\npageNum\nint\nOptional\nDefault to 1\n\nPage number\n\nValue range: [1, 1000]\npageSize\nint\nOptional\nDefault to 30\n\nNumber of items per page\n\nValue range: [1, 500]\nPrevious chapter：Models\nNext chapter：Image Generation\nCreate Task\nInvocation examples\nQuery Task (Single)\nQuery Task (List)\nThe Kling 3.0 Series Models API is Now Fully Available\n– All in One, One for All！\n\nModels Available in This Release\n\nKling 3.0 Motion Control, Kling Video 3.0, Kling Video 3.0 Omni, Kling Image 3.0, Kling Image 3.0 Omni\n\nRefer to \n\nKey Highlights of the Models\n\n3.0 All-in-One: A unified model for multi-modal input and output.\n\nMost powerful consistency across the universe: Subject consistency (supports cameo, subject with voice control, i2v + subject) and text consistency.\nNarrative control at your fingertips: More freedom, precision, and control—up to 15 seconds long, video scene cuts, ultra-high-definition storyboards/images, custom seconds.\nUpgraded native audio-visual output: Supports multiple speakers and languages (with accents).\n\nKling 3.0 Motion Control\n\nConsistent Facial Identity from any angle\nComplex Emotions faithfully repr", "Image Generation": "Motion Control\nMulti-elements to video\nExtend Video\nLip Sync\nAvatar\nText to Audio\nVideo to Audio\nText to Speech\nVoice Clone\nImage Recognize\nElement\nEffects\nEffect Templates\nNEW\nVideo Effects\nImage Generation\nModels\nImage Omni\nImage Generation\nReference to Image\nExtend Image\nAI Multi-Shot\nVirtual Try-On\nOthers\nQuery user info\nPricing\nBilling Info\nPrepaid Resource Packs\nProtocols\nPrivacy Policy of API Service\nTerms of API Service\nAPI Service Level Agreement\nImage Generation\nCreate Task\nPOST\n/v1/images/generations\ncURL\nCopy\nCollapse\ncurl --request POST \\\n --url https://api-singapore.klingai.com/v1/images/generations \\\n --header 'Authorization: Bearer ' \\\n --header 'Content-Type: application/json' \\\n --data '{\n \"model_name\": \"kling-v2-1\",\n \"prompt\": \"Generate a Pixar-style puppy\",\n \"negative_prompt\": \"\",\n \"image\": \"https://p1-kling.klingai.com/kcdn/cdn-kcdn112452/kling-qa-test/dog.png\",\n \"n\": 2,\n \"external_task_id\": \"\",\n \"callback_url\": \"\"\n}'\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes; specific definitions see Error codes\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system, for tracking and troubleshooting\n \"data\": {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_status\": \"string\", // Task status: submitted, processing, succeed, failed\n \"task_info\": { // Task creation parameters\n \"external_task_id\": \"string\" // Customer-defined task ID\n },\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, ms\n \"updated_at\": 1722769557708 // Task update time, Unix timestamp, ms\n }\n}\n💡\n\nPlease note that in order to maintain naming consistency, the original model field has been changed to model_name, so in the future, please use this field to specify the version of the model that needs to be called.\nAt the same time, we keep the behavior forward-compatible. If you continue to use the original model field, it will not have any impact on the interface call, there will not be any exception, which is equivalent to the default behavior when model_name is empty (i.e., call the V1 model).\n\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nRequest Body\nmodel_name\nstring\nOptional\nDefault to kling-v1\n\nModel Name\n\nEnum values：\nkling-v1\nkling-v1-5\nkling-v2\nkling-v2-new\nkling-v2-1\nkling-v3\nprompt\nstring\nRequired\n\nPositive text prompt\n\nCannot exceed 2500 characters\n\nnegative_prompt\nstring\nOptional\n\nNegative text prompt\n\nCannot exceed 2500 characters\n\nNote: In the Image-to-Image scenario (when the \"image\" field is not empty), negative prompts are not supported.\n\nimage\nstring\nOptional\n\nReference Image\n\nSupport inputting image Base64 encoding or image URL (ensure accessibility)\n\nBase64 Encoding Note:\nPlease note, if you use the Base64 method, make sure all image data parameters you pass are in Base64 encoding format. When using Base64, do NOT add any prefix like data:image/png;base64,. Only provide the Base64-encoded string.\n\nCorrect:\n\niVBORw0KGgoAAAANSUhEUgAAAAUA...\n\nIncorrect:\n\ndata:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAUA...\nSupported image formats: .jpg / .jpeg / .png\nFile size: ≤10MB, dimensions: min 300px, aspect ratio: 1:2.5 ~ 2.5:1\nRequired when image_reference parameter is not empty\nimage_reference\nstring\nOptional\n\nImage reference type\n\nEnum values：\nsubject\nface\nsubject: character feature reference, face: character appearance reference\nWhen using face, the uploaded image must contain only one face\n\nRequired when using kling-v1-5 and image parameter is not empty\n\nimage_fidelity\nfloat\nOptional\nDefault to 0.5\n\nFace reference intensity for user-uploaded images during generation\n\nValue range: [0, 1], The larger the value, the stronger the reference intensity\n\nOnly kling-v1, kling-v1-5 support this parameter\n\nhuman_fidelity\nfloat\nOptional\nDefault to 0.45\n\nFacial reference intensity, refers to the similarity of the facial features of the person in the reference image\n\nOnly image_reference parameter is subject is available\n\nValue range: [0, 1], The larger the value, the stronger the reference intensity\n\nOnly kling-v1-5 supports this parameter\n\nelement_list\narray\nOptional\n\nReference element list based on element library ID\n\nLoad with key:value format as follows:\n\"element_list\":[\n { \"element_id\": long },\n { \"element_id\": long }\n]\n\nThe amount of reference element is related to the amount of reference image, the sum of the amount of reference element and the amount of reference image shall not exceed 10.\n\n▾\nHide child attributes\nelement_id\nlong\nRequired\n\nElement ID\n\nresolution\nstring\nOptional\nDefault to 1k\n\nImage generation resolution\n\nEnum values：\n1k\n2k\n1k: 1K standard, 2k: 2K high-res\n\nDifferent model versions support varying ranges. For details, refer to the Capability Map\n\nn\nint\nOptional\nDefault to 1\n\nNumber of generated images\n\nValue range: [1, 9]\n\naspect_ratio\nstring\nOptional\nDefault to 16:9\n\nAspect ratio of the generated images (width:height)\n\nEnum values：\n16:9\n9:16\n1:1\n4:3\n3:4\n3:2\n2:3\n21:9\n\nDifferent model versions support varying ranges. For details, refer to the Capability Map\n\nwatermark_info\nobject\nOptional\n\nWhether to generate watermarked results simultaneously\n\nDefined by the enabled parameter, format:\n \"watermark_info\": { \"enabled\": boolean } \ntrue: generate watermarked result, false: do not generate\nCustom watermarks are not currently supported\ncallback_url\nstring\nOptional\n\nThe callback notification address for the result of this task. If configured, the server will actively notify when the task status changes.\n\nFor specific message schema, see Callback Protocol\nexternal_task_id\nstring\nOptional\n\nCustomized Task ID\n\nWill not overwrite system-generated task ID, but supports querying task by this ID\nPlease note that the customized task ID must be unique within a single user account.\nQuery Task (Single)\nGET\n/v1/images/generations/{id}\ncURL\nCopy\nCollapse\ncurl --request GET \\\n --url https://api-singapore.klingai.com/v1/images/generations/{id} \\\n --header 'Authorization: Bearer '\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes; Specific definitions can be found in Error codes\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system, is used to track requests and troubleshoot problems\n \"data\": {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_status\": \"string\", // Task status, Enum values: submitted, processing, succeed, failed\n \"task_status_msg\": \"string\", // Task status information, displaying the failure reason when the task fails (such as triggering the content risk control of the platform, etc.)\n \"final_unit_deduction\": \"string\", // The deduction units of task\n \"watermark_info\": {\n \"enabled\": boolean\n },\n \"task_info\": { // Task creation parameters\n \"external_task_id\": \"string\" // Customer-defined task ID\n },\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit ms\n \"updated_at\": 1722769557708, // Task update time, Unix timestamp, unit ms\n \"task_result\": {\n \"images\": [\n {\n \"index\": 0, // Image Number, 0-9\n \"url\": \"string\", // URL for generating images, such as: https://h1.inkwai.com/bs2/upload-ylab-stunt/1fa0ac67d8ce6cd55b50d68b967b3a59.png(To ensure information security, generated images/videos will be cleared after 30 days. Please make sure to save them promptly.)\n \"watermark_url\": \"string\" // Watermarked image download URL, anti-leech format\n }\n ]\n }\n }\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nPath Parameters\ntask_id\nstring\nRequired\n\nThe task ID for image generation. Request path parameter, directly fill the value in the request path.\n\nexternal_task_id\nstring\nOptional\n\nCustomized Task ID for audio generation\n\nThe external_task_id filled in when creating the task. You can choose to query by external_task_id or task_id\nWhen creating a task, you can choose to query by external_task_id or task_id.\nQuery Task (List)\nGET\n/v1/images/generations\ncURL\nCopy\nCollapse\ncurl --request GET \\\n --url 'https://api-singapore.klingai.com/v1/images/generations?pageNum=1&pageSize=30' \\\n --header 'Authorization: Bearer '\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes; Specific definitions can be found in Error codes\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system, is used to track requests and troubleshoot problems\n \"data\": [\n {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_status\": \"string\", // Task status, Enum values: submitted, processing, succeed, failed\n \"task_status_msg\": \"string\", // Task status information, displaying the failure reason when the task fails (such as triggering the content risk control of the platform, etc.)\n \"final_unit_deduction\": \"string\", // The deduction units of task\n \"watermark_info\": {\n \"enabled\": boolean\n },\n \"task_info\": { // Task creation parameters\n \"external_task_id\": \"string\" // Customer-defined task ID\n },\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit ms\n \"updated_at\": 1722769557708, // Task update time, Unix timestamp, unit ms\n \"task_result\": {\n \"images\": [\n {\n \"index\": 0, // Image Number, 0-9\n \"url\": \"string\", // URL for generating images, such as: https://h1.inkwai.com/bs2/upload-ylab-stunt/1fa0ac67d8ce6cd55b50d68b967b3a59.png(To ensure information security, generated images/videos will be cleared after 30 days. Please make sure to save them promptly.)\n \"watermark_url\": \"string\" // Watermarked image download URL, anti-leech format\n }\n ]\n }\n }\n ]\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nQuery Parameters\npageNum\nint\nOptional\nDefault to 1\n\nPage number\n\nValue range: [1, 1000]\n\npageSize\nint\nOptional\nDefault to 30\n\nNumber of items per page\n\nValue range: [1, 500]\n\nPrevious chapter：Image Omni\nNext chapter：Reference to Image\nCreate Task\nQuery Task (Single)\nQuery Task (List)\nThe Kling 3.0 Series Models API is Now Fully Available\n– All in One, One for All！\n\nModels Available in This Release\n\nKling 3.0 Motion Control, Kling Video 3.0, Kling Video 3.0 Omni, Kling Image 3.0, Kling Image 3.0 Omni\n\nRefer to \n\nKey Highlights of the Models\n\n3.0 All-in-One: A unified model for multi-modal input and output.\n\nMost powerful consistency across the universe: Subject consistency (supports cameo, subject with voice control, i2v + subject) and text consistency.\nNarrative control at your fingertips: More freedom, precision, and control—up to 15 seconds long, video scene cuts, ultra-high-definition storyboards/images, custom seconds.\nUpgraded native audio-visual output: Supports multiple speakers and languages (with accents).\n\nKling 3.0 Motion Control\n\nConsistent Facial Identity from any angle\nComplex Emotions faithfully reproduced\nHigh fidelity Restoration, Even with Face Occlusions\nConsistent Facial Clarity Across Dynamic Framing\n\nUser Guide ->\n\nKling Video 3.0\n\nCompared to 2.6, expected improvements:\n\nSupports subject upload in I2V scenarios for enhanced consistency\nSignificant improvement in multi-character referencing, especially for three-person scenarios\nSupports Japanese, Korean, and Spanish in addition to Chinese and English\nCapable of generating certain dialects and accents\nBetter distinction and control over different types of audio (speech, sound effects, BGM)\nImproved text retention in I2V scenarios\nSupports scene transitions, ", "Reference to Image": "Motion Control\nMulti-elements to video\nExtend Video\nLip Sync\nAvatar\nText to Audio\nVideo to Audio\nText to Speech\nVoice Clone\nImage Recognize\nElement\nEffects\nEffect Templates\nNEW\nVideo Effects\nImage Generation\nModels\nImage Omni\nImage Generation\nReference to Image\nExtend Image\nAI Multi-Shot\nVirtual Try-On\nOthers\nQuery user info\nPricing\nBilling Info\nPrepaid Resource Packs\nProtocols\nPrivacy Policy of API Service\nTerms of API Service\nAPI Service Level Agreement\nMulti-Image to Image\nCreate Task\nPOST\n/v1/images/multi-image2image\ncURL\nCopy\nCollapse\ncurl --request POST \\\n --url https://api-singapore.klingai.com/v1/images/multi-image2image \\\n --header 'Authorization: Bearer ' \\\n --header 'Content-Type: application/json' \\\n --data '{\n \"model_name\": \"kling-v2-1\",\n \"prompt\": \"Wearing a flowing red dress on the grassland, in Ghibli style.\",\n \"negative_prompt\": \"\",\n \"subject_image_list\": [\n { \"subject_image\": \"https://v1-kling.klingai.com/kcdn/cdn-kcdn112452/kling-qa-test/multi-1.png\" },\n { \"subject_image\": \"https://v1-kling.klingai.com/kcdn/cdn-kcdn112452/kling-qa-test/multi-2.png\" }\n ],\n \"scene_image\": \"https://v1-kling.klingai.com/kcdn/cdn-kcdn112452/kling-qa-test/background.jpeg\",\n \"style_image\": \"https://v1-kling.klingai.com/kcdn/cdn-kcdn112452/kling-qa-test/16x9_jipuli.png\",\n \"n\": 2,\n \"aspect_ratio\": \"9:16\"\n }'\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error code; Specific definitions can be found in \"Error Code\"\n \"message\": \"string\", // Error message\n \"request_id\": \"string\", // Request ID, generated by the system, used for tracking requests and troubleshooting\n \"data\": {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_info\": { //Task creation parameters\n \"external_task_id\": \"string\" //User-defined task ID\n },\n \"task_status\": \"string\", // Task status, Enum values: submitted, processing, succeed, failed\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit: ms\n \"updated_at\": 1722769557708 //Task update time, Unix timestamp, unit: ms\n }\n}\n\nGenerate images based on multiple reference images (subject, scene, style).\n\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nRequest Body\nmodel_name\nstring\nOptional\nDefault to kling-v2\n\nModel Name\n\nEnum values：\nkling-v2\nkling-v2-1\nprompt\nstring\nOptional\n\nPositive text prompt\n\nCannot exceed 2500 characters\nsubject_image_list\narray\nRequired\n\nSubject Reference Images\n\nSupport up to 4 images, at least 1 image, using key:value format as follows:\n\"subject_image_list\":[\n { \"subject_image\":\"image_url\" },\n { \"subject_image\":\"image_url\" },\n { \"subject_image\":\"image_url\" },\n { \"subject_image\":\"image_url\" }\n]\nThe API does not perform cropping, please upload images with subjects already cropped\nSupports image input as either Base64-encoded string or URL (ensure the URL is publicly accessible)\nImportant: When using Base64, do NOT add any prefix like data:image/png;base64,. Submit only the raw Base64 string.\nCorrect Base64 format:\niVBORw0KGgoAAAANSUhEUgAAAAUA...\nIncorrect Base64 format (with data: prefix):\ndata:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAUA...\nSupported image formats: .jpg / .jpeg / .png\nImage file size must not exceed 10MB. Image dimensions must be at least 300px in width and height. Aspect ratio must be between 1:2.5 and 2.5:1\n▾\nHide child attributes\nsubject_image\nstring\nRequired\n\nSubject image URL or Base64 string\n\nscene_image\nstring\nOptional\n\nScene Reference Image\n\nSupports image input as either Base64-encoded string or URL (ensure the URL is publicly accessible)\nImportant: When using Base64, do NOT add any prefix like data:image/png;base64,. Submit only the raw Base64 string.\nCorrect Base64 format:\niVBORw0KGgoAAAANSUhEUgAAAAUA...\nIncorrect Base64 format (with data: prefix):\ndata:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAUA...\nSupported image formats: .jpg / .jpeg / .png\nImage file size must not exceed 10MB. Image dimensions must be at least 300px in width and height. Aspect ratio must be between 1:2.5 and 2.5:1\nstyle_image\nstring\nOptional\n\nStyle Reference Image\n\nSupports image input as either Base64-encoded string or URL (ensure the URL is publicly accessible)\nImportant: When using Base64, do NOT add any prefix like data:image/png;base64,. Submit only the raw Base64 string.\nCorrect Base64 format:\niVBORw0KGgoAAAANSUhEUgAAAAUA...\nIncorrect Base64 format (with data: prefix):\ndata:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAUA...\nSupported image formats: .jpg / .jpeg / .png\nImage file size must not exceed 10MB. Image dimensions must be at least 300px in width and height. Aspect ratio must be between 1:2.5 and 2.5:1\nn\nint\nOptional\nDefault to 1\n\nNumber of generated images\n\nValue range: [1, 9]\naspect_ratio\nstring\nOptional\nDefault to 16:9\n\nAspect ratio of the generated images (width:height)\n\nEnum values：\n16:9\n9:16\n1:1\n4:3\n3:4\n3:2\n2:3\n21:9\n\nThe support range for different model versions varies. For details, see Capability Map\n\nwatermark_info\nobject\nOptional\n\nWhether to generate watermarked results simultaneously\n\nDefined by the enabled parameter, format:\n \"watermark_info\": { \"enabled\": boolean } \ntrue: generate watermarked result, false: do not generate\nCustom watermarks are not currently supported\ncallback_url\nstring\nOptional\n\nThe callback notification address for the result of this task. If configured, the server will actively notify when the task status changes\n\nThe specific message schema of the notification can be found in Callback Protocol\nexternal_task_id\nstring\nOptional\n\nCustomized Task ID\n\nUser-defined task ID. It will not override the system-generated task ID, but supports querying tasks by this ID\nPlease note that it must be unique for each user\nQuery Task (Single)\nGET\n/v1/images/multi-image2image/{id}\ncURL\nCopy\nCollapse\ncurl --request GET \\\n --url https://api-singapore.klingai.com/v1/images/multi-image2image/{id} \\\n --header 'Authorization: Bearer '\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes; specific definitions see Error codes\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system\n \"data\": {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_info\": { //Task creation parameters\n \"external_task_id\": \"string\" //Customer-defined task ID\n },\n \"task_status\": \"string\", // Task status: submitted, processing, succeed, failed\n \"task_status_msg\": \"string\", // Task status message, failure reason when task fails\n \"final_unit_deduction\": \"string\", // Final unit deduction for the task\n \"watermark_info\": { \"enabled\": boolean },\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit: ms\n \"updated_at\": 1722769557708, // Task update time, Unix timestamp, unit: ms\n \"task_result\": {\n \"images\": [\n {\n \"index\": 0, // Image number, 0-9\n \"url\": \"string\", // URL for generated image (Generated images/videos will be cleared after 30 days. Please save promptly.)\n \"watermark_url\": \"string\" // Watermarked image download URL, anti-hotlinking format\n }\n ]\n }\n }\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nPath Parameters\ntask_id\nstring\nRequired\n\nThe task ID for image generation. Request path parameter, directly fill the value in the request path. Query by external_task_id or task_id.\n\nexternal_task_id\nstring\nOptional\n\nCustomized Task ID for audio generation\n\nThe external_task_id filled in when creating the task. You can choose to query by external_task_id or task_id\nWhen creating a task, you can choose to query by external_task_id or task_id.\nQuery Task (List)\nGET\n/v1/images/multi-image2image\ncURL\nCopy\nCollapse\ncurl --request GET \\\n --url 'https://api-singapore.klingai.com/v1/images/multi-image2image?pageNum=1&pageSize=30' \\\n --header 'Authorization: Bearer '\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes; specific definitions see Error codes\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system\n \"data\": [\n {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_info\": { //Task creation parameters\n \"external_task_id\": \"string\" //Customer-defined task ID\n },\n \"task_status\": \"string\", // Task status: submitted, processing, succeed, failed\n \"task_status_msg\": \"string\", // Task status message\n \"final_unit_deduction\": \"string\", // Final unit deduction for the task\n \"watermark_info\": { \"enabled\": true },\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit: ms\n \"updated_at\": 1722769557708, // Task update time, Unix timestamp, unit: ms\n \"task_result\": {\n \"images\": [\n {\n \"index\": 0, // Image number, 0-9\n \"url\": \"string\", // URL for generated image (Generated images/videos will be cleared after 30 days. Please save promptly.)\n \"watermark_url\": \"string\" // Watermarked image download URL, anti-hotlinking format\n }\n ]\n }\n }\n ]\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nQuery Parameters\npageNum\nint\nOptional\nDefault to 1\n\nPage number\n\nValue range: [1, 1000]\npageSize\nint\nOptional\nDefault to 30\n\nNumber of items per page\n\nValue range: [1, 500]\nPrevious chapter：Image Generation\nNext chapter：Extend Image\nCreate Task\nQuery Task (Single)\nQuery Task (List)\nThe Kling 3.0 Series Models API is Now Fully Available\n– All in One, One for All！\n\nModels Available in This Release\n\nKling 3.0 Motion Control, Kling Video 3.0, Kling Video 3.0 Omni, Kling Image 3.0, Kling Image 3.0 Omni\n\nRefer to \n\nKey Highlights of the Models\n\n3.0 All-in-One: A unified model for multi-modal input and output.\n\nMost powerful consistency across the universe: Subject consistency (supports cameo, subject with voice control, i2v + subject) and text consistency.\nNarrative control at your fingertips: More freedom, precision, and control—up to 15 seconds long, video scene cuts, ultra-high-definition storyboards/images, custom seconds.\nUpgraded native audio-visual output: Supports multiple speakers and languages (with accents).\n\nKling 3.0 Motion Control\n\nConsistent Facial Identity from any angle\nComplex Emotions faithfully reproduced\nHigh fidelity Restoration, Even with Face Occlusions\nConsistent Facial Clarity Across Dynamic Framing\n\nUser Guide ->\n\nKling Video 3.0\n\nCompared to 2.6, expected improvements:\n\nSupports subject upload in I2V scenarios for enhanced consistency\nSignificant improvement in multi-character referencing, especially for three-person scenarios\nSupports Japanese, Korean, and Spanish in addition to Chinese and English\nCapable of generating certain dialects and accents\nBetter distinction and control over different types of audio (speech, sound effects, BGM)\nImproved text retention in I2V scenarios\nSupports scene transitions, with up to 6 shots and customizable storyboarding\n\nUser Guide ->\n\nKling Video 3.0 Omni\n\nCompared to O1, expected improvements:\n\nNative audio-visual synchronization\nSupports video subject creation\nFurther improved consistency in reference-based tasks, especially for characters and products\nCombined capabilities of reference + storyboarding + audio-visual sync significantly enhance usability\nSupports scene transitions, with up to 6 shots\nExtended generation duration up to 15 seconds\n\nUser Guide ->\n\nKling Image 3.0\n\nHighly consistent feature retention\nPrecise response to detail modifications\nAccurate control over style and tone\nRich imaginative capabilities\n\nUser Guide ->\n\nKling Image 3.0 Omni\n\nEnhanced narrative sens", "Extend Image": "Motion Control\nMulti-elements to video\nExtend Video\nLip Sync\nAvatar\nText to Audio\nVideo to Audio\nText to Speech\nVoice Clone\nImage Recognize\nElement\nEffects\nEffect Templates\nNEW\nVideo Effects\nImage Generation\nModels\nImage Omni\nImage Generation\nReference to Image\nExtend Image\nAI Multi-Shot\nVirtual Try-On\nOthers\nQuery user info\nPricing\nBilling Info\nPrepaid Resource Packs\nProtocols\nPrivacy Policy of API Service\nTerms of API Service\nAPI Service Level Agreement\nImage Expansion\nCreate Task\nPOST\n/v1/images/editing/expand\ncURL\nCopy\nCollapse\ncurl --request POST \\\n --url https://api-singapore.klingai.com/v1/images/editing/expand \\\n --header 'Authorization: Bearer ' \\\n --header 'Content-Type: application/json' \\\n --data '{\n \"up_expansion_ratio\": 0.1495,\n \"down_expansion_ratio\": 0.1495,\n \"left_expansion_ratio\": 0.6547,\n \"right_expansion_ratio\": 0.6547,\n \"prompt\": \"\",\n \"image\": \"https://p1-kling.klingai.com/kcdn/cdn-kcdn112452/kling-qa-test/dog.png\",\n \"n\": 2,\n \"external_task_id\": \"\"\n }'\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error code; Specific definitions can be found in Error code\n \"message\": \"string\", // Error message\n \"request_id\": \"string\", // Request ID, generated by the system, is used to track requests and troubleshoot problems\n \"data\": {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_info\": { // Task creation parameters\n \"external_task_id\": \"string\" // Customer-defined task ID\n },\n \"task_status\": \"string\", // Task status, Enum values: submitted, processing, succeed, failed\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit: ms\n \"updated_at\": 1722769557708 // Task update time, Unix timestamp, unit: ms\n }\n}\n\nExpand the image in any direction based on the original image.\n\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nRequest Body\nimage\nstring\nRequired\n\nReference Image\n\nSupports image input as either Base64-encoded string or URL (ensure accessibility)\nImportant: When using Base64, do NOT add any prefix like data:image/png;base64,. Submit only the raw Base64 string.\nCorrect Base64 format:\niVBORw0KGgoAAAANSUhEUgAAAAUA...\nIncorrect Base64 format (with data: prefix):\ndata:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAUA...\nSupported image formats: .jpg / .jpeg / .png\nImage file size must not exceed 10MB. Image dimensions must be at least 300px in width and height, aspect ratio: 1:2.5 ~ 2.5:1\nup_expansion_ratio\nfloat\nRequired\n\nExpand upwards range; calculated based on multiples of the original image height\n\nValue range: [0, 2]. The total area of the new image must not exceed 3 times that of the original image\nExample: If original height is 20 and value is 0.1, then:\nThe distance from the top edge of the original image to the top edge of the new image is 20 × 0.1 = 2, and the area within is the expansion range\ndown_expansion_ratio\nfloat\nRequired\n\nExpand downwards range; calculated based on multiples of the original image height\n\nValue range: [0, 2]. The total area of the new image must not exceed 3 times that of the original image\nExample: If original height is 20 and value is 0.2, then:\nThe distance from the bottom edge of the original image to the bottom edge of the new image is 20 × 0.2 = 4, and the area within is the expansion range\nleft_expansion_ratio\nfloat\nRequired\n\nExpand leftwards range; calculated based on multiples of the original image width\n\nValue range: [0, 2]. The total area of the new image must not exceed 3 times that of the original image\nExample: If original width is 30 and value is 0.3, then:\nThe distance from the left edge of the original image to the left edge of the new image is 30 × 0.3 = 9, and the area within is the expansion range\nright_expansion_ratio\nfloat\nRequired\n\nExpand rightwards range; calculated based on multiples of the original image width\n\nValue range: [0, 2]. The total area of the new image must not exceed 3 times that of the original image\nExample: If original width is 30 and value is 0.4, then:\nThe distance from the right edge of the original image to the right edge of the new image is 30 × 0.4 = 12, and the area within is the expansion range\nprompt\nstring\nOptional\n\nPositive text prompt\n\nCannot exceed 2500 characters\nn\nint\nOptional\nDefault to 1\n\nNumber of generated images\n\nValue range: [1, 9]\nwatermark_info\nobject\nOptional\n\nWhether to generate watermarked results simultaneously\n\nDefined by the enabled parameter, format:\n \"watermark_info\": { \"enabled\": boolean } \ntrue: generate watermarked result, false: do not generate\nCustom watermarks are not currently supported\ncallback_url\nstring\nOptional\n\nThe callback notification address for the result of this task. If configured, the server will actively notify when the task status changes.\n\nFor specific message schema, see Callback Protocol\nexternal_task_id\nstring\nOptional\n\nCustomized Task ID\n\nWill not overwrite system-generated task ID, but supports querying task by this ID\nMust be unique within a single user account\nExample code\nimport math\n\ndef calculate_expansion_ratios(width, height, area_multiplier, aspect_ratio):\n \"\"\"\n Calculate top/bottom/left/right expansion ratios for image outpainting.\n \n Parameters:\n - width: Original image width\n - height: Original image height\n - area_multiplier: Multiplier for the outpainted area relative to original image\n - aspect_ratio: Width/height ratio for the outpainted area（width/height）\n \n Returns:\n - Formatted string with 4 decimal places, e.g.\"0.1495,0.1495,0.6547,0.6547\"\n \"\"\"\n # Calculate target total area\n target_area = area_multiplier * width * height\n \n # Calculate target height and width (maintaining aspect ratio)\n target_height = math.sqrt(target_area / aspect_ratio)\n target_width = target_height * aspect_ratio\n \n # Calculate expansion pixels\n expand_top = (target_height - height) / 2\n expand_bottom = expand_top\n expand_left = (target_width - width) / 2\n expand_right = expand_left\n \n # Calculate relative ratios\n top_ratio = expand_top / height\n bottom_ratio = expand_bottom / height\n left_ratio = expand_left / width\n right_ratio = expand_right / width\n \n # Format to 4 decimal places\n return f\"{top_ratio:.4f},{bottom_ratio:.4f},{left_ratio:.4f},{right_ratio:.4f}\"\n\n# Example: Original 100x100, 3x area multiplier, 16:9 aspect ratio\nprint(calculate_expansion_ratios(100, 100, 3, 16/9)) \n# Output: \"0.1495,0.1495,0.6547,0.6547\"\nQuery Task (Single)\nGET\n/v1/images/editing/expand/{id}\ncURL\nCopy\nCollapse\ncurl --request GET \\\n --url https://api-singapore.klingai.com/v1/images/editing/expand/{id} \\\n --header 'Authorization: Bearer '\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error code; Specific definitions can be found in Error code\n \"message\": \"string\", // Error message\n \"request_id\": \"string\", // Request ID, generated by the system, is used to track requests and troubleshoot problems\n \"data\": [\n {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_status\": \"string\", // Task status, Enum values: submitted, processing, succeed, failed\n \"task_status_msg\": \"string\", // Task status message, displaying the failure reason when the task fails (such as triggering the content risk control of the platform, etc.)\n \"task_info\": { // Task creation parameters\n \"external_task_id\": \"string\" // Customer-defined task ID\n },\n \"final_unit_deduction\": \"string\", // Final unit deduction for the task\n \"watermark_info\": {\n \"enabled\": boolean\n },\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit: ms\n \"updated_at\": 1722769557708, // Task update time, Unix timestamp, unit: ms\n \"task_result\": {\n \"images\": [\n {\n \"index\": 0, // Image index, 0-9\n \"url\": \"string\", // URL for generating images (To ensure information security, generated images/videos will be cleared after 30 days. Please make sure to save them promptly.)\n }\n ]\n }\n }\n ]\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nPath Parameters\ntask_id\nstring\nRequired\n\nThe task ID generated by images. Request path parameter, directly fill the value in the request path.\n\nexternal_task_id\nstring\nOptional\n\nCustomized Task ID for audio generation\n\nThe external_task_id filled in when creating the task. You can choose to query by external_task_id or task_id\nWhen creating a task, you can choose to query by external_task_id or task_id.\nQuery Task (List)\nGET\n/v1/images/editing/expand\ncURL\nCopy\nCollapse\ncurl --request GET \\\n --url 'https://api-singapore.klingai.com/v1/images/editing/expand?pageNum=1&pageSize=30' \\\n --header 'Authorization: Bearer '\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error code; Specific definitions can be found in Error code\n \"message\": \"string\", // Error message\n \"request_id\": \"string\", // Request ID, generated by the system, is used to track requests and troubleshoot problems\n \"data\": [\n {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_status\": \"string\", // Task status, Enum values: submitted, processing, succeed, failed\n \"task_status_msg\": \"string\", // Task status message, displaying the failure reason when the task fails (such as triggering the content risk control of the platform, etc.)\n \"task_info\": { // Task creation parameters\n \"external_task_id\": \"string\" // Customer-defined task ID\n },\n \"final_unit_deduction\": \"string\", // Final unit deduction for the task\n \"watermark_info\": {\n \"enabled\": boolean\n },\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit: ms\n \"updated_at\": 1722769557708, // Task update time, Unix timestamp, unit: ms\n \"task_result\": {\n \"images\": [\n {\n \"index\": 0, // Image index, 0-9\n \"url\": \"string\", // URL for generating images (To ensure information security, generated images/videos will be cleared after 30 days. Please make sure to save them promptly.)\n \"watermark_url\": \"string\", // Watermarked image download URL, anti-hotlinking format\n }\n ]\n }\n }\n ]\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nQuery Parameters\npageNum\nint\nOptional\nDefault to 1\n\nPage number\n\nValue range: [1, 1000]\npageSize\nint\nOptional\nDefault to 30\n\nData volume per page\n\nValue range: [1, 500]\nPrevious chapter：Reference to Image\nNext chapter：AI Multi-Shot\nCreate Task\nExample code\nQuery Task (Single)\nQuery Task (List)\nThe Kling 3.0 Series Models API is Now Fully Available\n– All in One, One for All！\n\nModels Available in This Release\n\nKling 3.0 Motion Control, Kling Video 3.0, Kling Video 3.0 Omni, Kling Image 3.0, Kling Image 3.0 Omni\n\nRefer to \n\nKey Highlights of the Models\n\n3.0 All-in-One: A unified model for multi-modal input and output.\n\nMost powerful consistency across the universe: Subject consistency (supports cameo, subject with voice control, i2v + subject) and text consistency.\nNarrative control at your fingertips: More freedom, precision, and control—up to 15 seconds long, video scene cuts, ultra-high-definition storyboards/images, custom seconds.\nUpgraded native audio-visual output: Supports multiple speakers and languages (with accents).\n\nKling 3.0 Motion Control\n\nConsistent Facial Identity from any angle\nComplex Emotions faithfully reproduced\nHigh fidelity Restoration, Even with Face Occlusions\nConsistent Facial Clarity Across Dynamic Framing\n\nUser Guide ->\n\nKling Video 3.0\n\nCompared to 2.6, expected improvements:\n\nSupp", "AI Multi-Shot": "Motion Control\nMulti-elements to video\nExtend Video\nLip Sync\nAvatar\nText to Audio\nVideo to Audio\nText to Speech\nVoice Clone\nImage Recognize\nElement\nEffects\nEffect Templates\nNEW\nVideo Effects\nImage Generation\nModels\nImage Omni\nImage Generation\nReference to Image\nExtend Image\nAI Multi-Shot\nVirtual Try-On\nOthers\nQuery user info\nPricing\nBilling Info\nPrepaid Resource Packs\nProtocols\nPrivacy Policy of API Service\nTerms of API Service\nAPI Service Level Agreement\nAI Multi-Shot\nCreate Task\nPOST\n/v1/general/ai-multi-shot\ncURL\nCopy\nCollapse\ncurl --request POST \\\n --url https://api-singapore.klingai.com/v1/general/ai-multi-shot \\\n --header 'Authorization: Bearer ' \\\n --header 'Content-Type: application/json' \\\n --data '{\n \"element_frontal_image\": \"https://v1-kling.klingai.com/kcdn/cdn-kcdn112452/kling-qa-test/multi-1.png\",\n \"external_task_id\": \"\",\n \"callback_url\": \"\"\n }'\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error code; Specific definitions can be found in \"Error Code\"\n \"message\": \"string\", // Error message\n \"request_id\": \"string\", // Request ID, generated by the system, used to track requests and troubleshoot problems\n \"data\": {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_status\": \"string\", // Task status, Enum values: submitted, processing, succeed, failed\n \"task_info\": { //Task creation parameters\n \"external_task_id\": \"string\" //Customer-defined task ID\n },\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit: ms\n \"updated_at\": 1722769557708 //Task update time, Unix timestamp, unit: ms\n }\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nRequest Body\nelement_frontal_image\nstring\nRequired\n\nThe frontal side of element image\n\nSupports inputting image Base64 encoding or image URL (ensure accessibility).\nSupported image formats: .jpg / .jpeg / .png\nImage file size cannot exceed 10MB, and the width and height dimensions shall not be less than 300px, and the aspect ratio should be between 1:2.5 ~ 2.5:1\ncallback_url\nstring\nOptional\n\nThe callback notification address for the result of this task. If configured, the server will actively notify when the task status changes\n\nFor the specific message schema, see Callback Protocol\nexternal_task_id\nstring\nOptional\n\nCustomized Task ID\n\nUsers can provide a customized task ID, which will not overwrite the system-generated task ID but can be used for task queries\nPlease note that the customized task ID must be unique within a single user account\nQuery Task (Single)\nGET\n/v1/general/ai-multi-shot/{id}\ncURL\nCopy\nCollapse\ncurl --request GET \\\n --url https://api-singapore.klingai.com/v1/general/ai-multi-shot/{task_id} \\\n --header 'Authorization: Bearer '\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error code; Specific definitions can be found in \"Error Code\"\n \"message\": \"string\", // Error message\n \"request_id\": \"string\", // Request ID, generated by the system, used to track requests and troubleshoot problems\n \"data\": {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_status\": \"string\", // Task status, Enum values: submitted, processing, succeed, failed\n \"task_status_msg\": \"string\", // Task status message, displaying the failure reason when the task fails (such as triggering the platform's content risk control, etc.)\n \"task_info\": { // Task creation parameters\n \"external_task_id\": \"string\" // Customer-defined task ID\n },\n \"task_result\": {\n \"images\": [\n {\n \"index\": 0, // Image result index, 0-2\n \"url_1\": \"string\", // URL for generated image, anti-hotlinking format (To ensure information security, generated images/videos will be cleared after 30 days. Please make sure to save them promptly.)\n \"url_2\": \"string\", // URL for generated image, anti-hotlinking format (To ensure information security, generated images/videos will be cleared after 30 days. Please make sure to save them promptly.)\n \"url_3\": \"string\" // URL for generated image, anti-hotlinking format (To ensure information security, generated images/videos will be cleared after 30 days. Please make sure to save them promptly.)\n }\n ]\n },\n \"final_unit_deduction\": \"string\", // Final unit deduction for the task\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit: ms\n \"updated_at\": 1722769557708 // Task update time, Unix timestamp, unit: ms\n }\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nPath Parameters\ntask_id\nstring\nRequired\n\nTask ID for AI Multi-Shot. Fill the value directly in the request path.\n\nQuery Task (List)\nGET\n/v1/general/ai-multi-shot\ncURL\nCopy\nCollapse\ncurl --request GET \\\n --url 'https://api-singapore.klingai.com/v1/general/ai-multi-shot?pageNum=1&pageSize=30' \\\n --header 'Authorization: Bearer '\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error code; Specific definitions can be found in \"Error Code\"\n \"message\": \"string\", // Error message\n \"request_id\": \"string\", // Request ID, generated by the system, used to track requests and troubleshoot problems\n \"data\": [\n {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_status\": \"string\", // Task status, Enum values: submitted, processing, succeed, failed\n \"task_status_msg\": \"string\", // Task status message, displaying the failure reason when the task fails (such as triggering the platform's content risk control, etc.)\n \"task_info\": { //Task creation parameters\n \"external_task_id\": \"string\" //Customer-defined task ID\n },\n \"task_result\": {\n \"images\": [\n {\n \"index\": 0, // Image result index, 0-2\n \"url_1\": \"string\", // URL for generated image, anti-hotlinking format (Please note that for information security, generated images/videos will be deleted after 30 days. Please save them promptly.)\n \"url_2\": \"string\", // URL for generated image, anti-hotlinking format (Please note that for information security, generated images/videos will be deleted after 30 days. Please save them promptly.)\n \"url_3\": \"string\" //URL for generated image, anti-hotlinking format (Please note that for information security, generated images/videos will be deleted after 30 days. Please save them promptly.)\n }\n ]\n },\n \"final_unit_deduction\": \"string\", // Final unit deduction for the task\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit: ms\n \"updated_at\": 1722769557708 //Task update time, Unix timestamp, unit: ms\n }\n ]\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nQuery Parameters\npageNum\nint\nOptional\nDefault to 1\n\nPage number\n\nValue range: [1, 1000]\npageSize\nint\nOptional\nDefault to 30\n\nNumber of items per page\n\nValue range: [1, 500]\nPrevious chapter：Extend Image\nNext chapter：Virtual Try-On\nCreate Task\nQuery Task (Single)\nQuery Task (List)\nThe Kling 3.0 Series Models API is Now Fully Available\n– All in One, One for All！\n\nModels Available in This Release\n\nKling 3.0 Motion Control, Kling Video 3.0, Kling Video 3.0 Omni, Kling Image 3.0, Kling Image 3.0 Omni\n\nRefer to \n\nKey Highlights of the Models\n\n3.0 All-in-One: A unified model for multi-modal input and output.\n\nMost powerful consistency across the universe: Subject consistency (supports cameo, subject with voice control, i2v + subject) and text consistency.\nNarrative control at your fingertips: More freedom, precision, and control—up to 15 seconds long, video scene cuts, ultra-high-definition storyboards/images, custom seconds.\nUpgraded native audio-visual output: Supports multiple speakers and languages (with accents).\n\nKling 3.0 Motion Control\n\nConsistent Facial Identity from any angle\nComplex Emotions faithfully reproduced\nHigh fidelity Restoration, Even with Face Occlusions\nConsistent Facial Clarity Across Dynamic Framing\n\nUser Guide ->\n\nKling Video 3.0\n\nCompared to 2.6, expected improvements:\n\nSupports subject upload in I2V scenarios for enhanced consistency\nSignificant improvement in multi-character referencing, especially for three-person scenarios\nSupports Japanese, Korean, and Spanish in addition to Chinese and English\nCapable of generating certain dialects and accents\nBetter distinction and control over different types of audio (speech, sound effects, BGM)\nImproved text retention in I2V scenarios\nSupports scene transitions, with up to 6 shots and customizable storyboarding\n\nUser Guide ->\n\nKling Video 3.0 Omni\n\nCompared to O1, expected improvements:\n\nNative audio-visual synchronization\nSupports video subject creation\nFurther improved consistency in reference-based tasks, especially for characters and products\nCombined capabilities of reference + storyboarding + audio-visual sync significantly enhance usability\nSupports scene transitions, with up to 6 shots\nExtended generation duration up to 15 seconds\n\nUser Guide ->\n\nKling Image 3.0\n\nHighly consistent feature retention\nPrecise response to detail modifications\nAccurate control over style and tone\nRich imaginative capabilities\n\nUser Guide ->\n\nKling Image 3.0 Omni\n\nEnhanced narrative sense\nNew storyboard image set generation, retaining reference image features with scene relevance\nDirect output of 2K/4K ultra-high-definition images\nFurther improved detail consistency\n\nUser Guide ->\n\nThank you for your support and understanding!\n\nI Got It", "Virtual Try-On": "Motion Control\nMulti-elements to video\nExtend Video\nLip Sync\nAvatar\nText to Audio\nVideo to Audio\nText to Speech\nVoice Clone\nImage Recognize\nElement\nEffects\nEffect Templates\nNEW\nVideo Effects\nImage Generation\nModels\nImage Omni\nImage Generation\nReference to Image\nExtend Image\nAI Multi-Shot\nVirtual Try-On\nOthers\nQuery user info\nPricing\nBilling Info\nPrepaid Resource Packs\nProtocols\nPrivacy Policy of API Service\nTerms of API Service\nAPI Service Level Agreement\nVirtual Try-On\nCreate Task\nPOST\n/v1/images/kolors-virtual-try-on\ncURL\nCopy\nCollapse\ncurl --request POST \\\n --url https://api-singapore.klingai.com/v1/images/kolors-virtual-try-on \\\n --header 'Authorization: Bearer ' \\\n --header 'Content-Type: application/json' \\\n --data '{\n \"model_name\": \"kolors-virtual-try-on-v1-5\",\n \"human_image\": \"https://p1-kling.klingai.com/kcdn/cdn-kcdn112452/kling-qa-test/tryon-human.webp\",\n \"cloth_image\": \"https://p1-kling.klingai.com/kcdn/cdn-kcdn112452/kling-qa-test/suite.webp\",\n \"callback_url\": \"\",\n \"external_task_id\": \"\"\n }'\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes; specific definitions see Error codes\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system, for tracking and troubleshooting\n \"data\": {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_status\": \"string\", // Task status: submitted, processing, succeed, failed\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, ms\n \"updated_at\": 1722769557708 // Task update time, Unix timestamp, ms\n }\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nRequest Body\nmodel_name\nstring\nOptional\nDefault to kolors-virtual-try-on-v1\n\nModel Name\n\nEnum values：\nkolors-virtual-try-on-v1\nkolors-virtual-try-on-v1-5\nhuman_image\nstring\nRequired\n\nReference human Image\n\nSupports inputting image Base64 encoding or image URL (ensure accessibility)\nImportant: When using Base64, do NOT add any prefix like data:image/png;base64,. Submit only the raw Base64 string.\nCorrect Base64 format:\niVBORw0KGgoAAAANSUhEUgAAAAUA...\nIncorrect Base64 format (with data: prefix):\ndata:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAUA...\nSupported image formats: .jpg / .jpeg / .png\nThe image file size cannot exceed 10MB, and the width and height dimensions of the image shall not be less than 300px\ncloth_image\nstring\nRequired\n\nReference clothing image\n\nSupport uploading clothing product images or clothing image with white background; Supports single clothing (upper, lower, and dress) try-on\nSupports inputting image Base64 encoding or image URL (ensure accessibility)\n\nBase64 Encoding Note:\nPlease note, if you use the Base64 method, make sure all image data parameters you pass are in Base64 encoding format. When using Base64, do NOT add any prefix like data:image/png;base64,. Only provide the Base64-encoded string.\n\nCorrect:\n\niVBORw0KGgoAAAANSUhEUgAAAAUA...\n\nIncorrect:\n\ndata:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAUA...\nSupported image formats: .jpg / .jpeg / .png\nFile size: ≤10MB, dimensions: min 300px\nThe kolors-virtual-try-on-v1-5 model not only supports single clothing input, but also supports \"upper + lower\" combination input:\nInput a single clothing image (upper, lower, or dress) → Generate a try-on image of the single item\nInput a combination clothing image (merge multiple items into one image with white background):\n\"upper + lower\" → Success\n\"upper + upper\" → Fails\n\"lower + lower\" → Fails\n\"dress + dress\" → Fails\n\"upper + dress\" → Fails\n\"lower + dress\" → Fails\n\ncallback_url\nstring\nOptional\n\nThe callback notification address for the result of this task. If configured, the server will actively notify when the task status changes.\n\nFor the specific message schema, see Callback Protocol\nexternal_task_id\nstring\nOptional\n\nCustomized Task ID\n\nWill not overwrite system-generated task ID, but supports querying task by this ID\nPlease note that the customized task ID must be unique within a single user account.\nQuery Task (Single)\nGET\n/v1/images/kolors-virtual-try-on/{id}\ncURL\nCopy\nCollapse\ncurl --request GET \\\n --url https://api-singapore.klingai.com/v1/images/kolors-virtual-try-on/{id} \\\n --header 'Authorization: Bearer ' \\\n --header 'Content-Type: application/json'\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes; specific definitions see Error codes\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system, for tracking and troubleshooting\n \"data\": {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_status\": \"string\", // Task status: submitted, processing, succeed, failed\n \"task_status_msg\": \"string\", // Task status information, displaying the failure reason when the task fails (e.g. content risk control)\n \"final_unit_deduction\": \"string\", // The deduction units of task\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, ms\n \"updated_at\": 1722769557708, // Task update time, Unix timestamp, ms\n \"task_info\": { // Task creation parameters\n \"external_task_id\": \"string\" // Customer-defined task ID\n },\n \"task_result\": {\n \"images\": [\n {\n \"index\": 0, // Image Number\n \"url\": \"string\" // URL for generating images (To ensure information security, generated images/videos will be cleared after 30 days. Please save them promptly.)\n }\n ]\n }\n }\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nPath Parameters\nid\nstring\nRequired\n\nTask ID for virtual try-on. Request path parameter, directly fill the value in the request path.\n\nexternal_task_id\nstring\nOptional\n\nCustomized Task ID for audio generation\n\nThe external_task_id filled in when creating the task. You can choose to query by external_task_id or task_id\nWhen creating a task, you can choose to query by external_task_id or task_id.\nQuery Task (List)\nGET\n/v1/images/kolors-virtual-try-on\ncURL\nCopy\nCollapse\ncurl --request GET \\\n --url 'https://api-singapore.klingai.com/v1/images/kolors-virtual-try-on?pageNum=1&pageSize=30' \\\n --header 'Authorization: Bearer ' \\\n --header 'Content-Type: application/json'\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes; specific definitions see Error codes\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system, for tracking and troubleshooting\n \"data\": [\n {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_status\": \"string\", // Task status: submitted, processing, succeed, failed\n \"task_status_msg\": \"string\", // Task status information, displaying the failure reason when the task fails (e.g. content risk control)\n \"final_unit_deduction\": \"string\", // The deduction units of task\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, ms\n \"updated_at\": 1722769557708, // Task update time, Unix timestamp, ms\n \"task_info\": { // Task creation parameters\n \"external_task_id\": \"string\" // Customer-defined task ID\n },\n \"task_result\": {\n \"images\": [\n {\n \"index\": 0, // Image Number\n \"url\": \"string\" // URL for generating images (To ensure information security, generated images/videos will be cleared after 30 days. Please save them promptly.)\n }\n ]\n }\n }\n ]\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nQuery Parameters\npageNum\nint\nOptional\nDefault to 1\n\nPage number\n\nValue range: [1, 1000]\n\npageSize\nint\nOptional\nDefault to 30\n\nData volume per page\n\nValue range: [1, 500]\n\nPrevious chapter：AI Multi-Shot\nNext chapter：Query user info\nCreate Task\nQuery Task (Single)\nQuery Task (List)\nThe Kling 3.0 Series Models API is Now Fully Available\n– All in One, One for All！\n\nModels Available in This Release\n\nKling 3.0 Motion Control, Kling Video 3.0, Kling Video 3.0 Omni, Kling Image 3.0, Kling Image 3.0 Omni\n\nRefer to \n\nKey Highlights of the Models\n\n3.0 All-in-One: A unified model for multi-modal input and output.\n\nMost powerful consistency across the universe: Subject consistency (supports cameo, subject with voice control, i2v + subject) and text consistency.\nNarrative control at your fingertips: More freedom, precision, and control—up to 15 seconds long, video scene cuts, ultra-high-definition storyboards/images, custom seconds.\nUpgraded native audio-visual output: Supports multiple speakers and languages (with accents).\n\nKling 3.0 Motion Control\n\nConsistent Facial Identity from any angle\nComplex Emotions faithfully reproduced\nHigh fidelity Restoration, Even with Face Occlusions\nConsistent Facial Clarity Across Dynamic Framing\n\nUser Guide ->\n\nKling Video 3.0\n\nCompared to 2.6, expected improvements:\n\nSupports subject upload in I2V scenarios for enhanced consistency\nSignificant improvement in multi-character referencing, especially for three-person scenarios\nSupports Japanese, Korean, and Spanish in addition to Chinese and English\nCapable of generating certain dialects and accents\nBetter distinction and control over different types of audio (speech, sound effects, BGM)\nImproved text retention in I2V scenarios\nSupports scene transitions, with up to 6 shots and customizable storyboarding\n\nUser Guide ->\n\nKling Video 3.0 Omni\n\nCompared to O1, expected improvements:\n\nNative audio-visual synchronization\nSupports video subject creation\nFurther improved consistency in reference-based tasks, especially for characters and products\nCombined capabilities of reference + storyboarding + audio-visual sync significantly enhance usability\nSupports scene transitions, with up to 6 shots\nExtended generation duration up to 15 seconds\n\nUser Guide ->\n\nKling Image 3.0\n\nHighly consistent feature retention\nPrecise response to detail modifications\nAccurate control over style and tone\nRich imaginative capabilities\n\nUser Guide ->\n\nKling Image 3.0 Omni\n\nEnhanced narrative sense\nNew storyboard image set generation, retaining reference image features with scene relevance\nDirect output of 2K/4K ultra-high-definition images\nFurther improved detail consistency\n\nUser Guide ->\n\nThank you for your support and understanding!\n\nI Got It", "Text to Audio": "Motion Control\nMulti-elements to video\nExtend Video\nLip Sync\nAvatar\nText to Audio\nVideo to Audio\nText to Speech\nVoice Clone\nImage Recognize\nElement\nEffects\nEffect Templates\nNEW\nVideo Effects\nImage Generation\nModels\nImage Omni\nImage Generation\nReference to Image\nExtend Image\nAI Multi-Shot\nVirtual Try-On\nOthers\nQuery user info\nPricing\nBilling Info\nPrepaid Resource Packs\nProtocols\nPrivacy Policy of API Service\nTerms of API Service\nAPI Service Level Agreement\nText to Audio\nCreate Task\nPOST\n/v1/audio/text-to-audio\ncURL\nCopy\nCollapse\ncurl --request POST \\\n --url https://api-singapore.klingai.com/v1/audio/text-to-audio \\\n --header 'Authorization: Bearer ' \\\n --header 'Content-Type: application/json' \\\n --data '{\n \"prompt\": \"Fireworks sound during Chinese New Year celebration\",\n \"duration\": 3,\n \"external_task_id\": \"\",\n \"callback_url\": \"\"\n }'\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes; Specific definitions can be found in \"Error Code\"\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system, used to track requests and troubleshoot problems\n \"data\": {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_info\": { // Task creation parameters\n \"external_task_id\": \"string\" // Customer-defined task ID\n },\n \"task_status\": \"string\", // Task status, Enum values: submitted, processing, succeed, failed\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit: ms\n \"updated_at\": 1722769557708 // Task update time, Unix timestamp, unit: ms\n }\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nRequest Body\nprompt\nstring\nRequired\n\nText prompt\n\nCannot exceed 200 characters\nduration\nfloat\nRequired\n\nGenerated audio duration\n\nValue range: 3.0s - 10.0s, supports one decimal place precision\nexternal_task_id\nstring\nOptional\n\nCustomized Task ID\n\nUsers can provide a customized task ID, which will not overwrite the system-generated task ID but can be used for task queries.\nPlease note that the customized task ID must be unique within a single user account.\ncallback_url\nstring\nOptional\n\nThe callback notification address for the result of this task. If configured, the server will actively notify when the task status changes\n\nThe specific message schema of the notification can be found in Callback Protocol\nQuery Task (Single)\nGET\n/v1/audio/text-to-audio/{id}\ncURL\nCopy\nCollapse\ncurl --request GET \\\n --url https://api-singapore.klingai.com/v1/audio/text-to-audio/{task_id} \\\n --header 'Authorization: Bearer ' \\\n --header 'Content-Type: application/json'\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes; Specific definitions can be found in \"Error Code\"\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system, used to track requests and troubleshoot problems\n \"data\": {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_status\": \"string\", // Task status, Enum values: submitted, processing, succeed, failed\n \"task_status_msg\": \"string\", // Task status message, displaying the failure reason when the task fails (such as triggering the platform's content risk control, etc.)\n \"task_info\": { // Task creation parameters\n \"external_task_id\": \"string\" // Customer-defined task ID\n },\n \"task_result\": {\n \"audios\": [\n {\n \"id\": \"string\", // Audio ID; globally unique\n \"url_mp3\": \"string\", // URL for generated audio in MP3 format (Please note that for information security, generated audios will be cleared after 30 days. Please save them promptly.)\n \"url_wav\": \"string\", // URL for generated audio in WAV format (Please note that for information security, generated audios will be cleared after 30 days. Please save them promptly.)\n \"duration_mp3\": \"string\", // Total duration of the audio in MP3 format, unit: s\n \"duration_wav\": \"string\" // Total duration of the audio in WAV format, unit: s\n }\n ]\n },\n \"final_unit_deduction\": \"string\", // The deduction units of task\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit: ms\n \"updated_at\": 1722769557708 // Task update time, Unix timestamp, unit: ms\n }\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nPath Parameters\ntask_id\nstring\nOptional\n\nThe task ID for audio generation\n\nRequest path parameter, fill the value directly in the request path\nYou can choose to query by external_task_id or task_id\nexternal_task_id\nstring\nOptional\n\nCustomized Task ID for audio generation\n\nThe external_task_id filled in when creating the task. You can choose to query by external_task_id or task_id\nQuery Task (List)\nGET\n/v1/audio/text-to-audio\ncURL\nCopy\nCollapse\ncurl --request GET \\\n --url 'https://api-singapore.klingai.com/v1/audio/text-to-audio?pageNum=1&pageSize=30' \\\n --header 'Authorization: Bearer ' \\\n --header 'Content-Type: application/json'\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes; Specific definitions can be found in \"Error Code\"\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system, used to track requests and troubleshoot problems\n \"data\": [\n {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_status\": \"string\", // Task status, Enum values: submitted, processing, succeed, failed\n \"task_status_msg\": \"string\", // Task status message, displaying the failure reason when the task fails (such as triggering the platform's content risk control, etc.)\n \"task_info\": { // Task creation parameters\n \"external_task_id\": \"string\" // Customer-defined task ID\n },\n \"task_result\": {\n \"audios\": [\n {\n \"id\": \"string\", // Audio ID; globally unique\n \"url_mp3\": \"string\", // URL for generated audio in MP3 format (Please note that for information security, generated audios will be cleared after 30 days. Please save them promptly.)\n \"url_wav\": \"string\", // URL for generated audio in WAV format (Please note that for information security, generated audios will be cleared after 30 days. Please save them promptly.)\n \"duration_mp3\": \"string\", // Total duration of the audio in MP3 format, unit: s\n \"duration_wav\": \"string\" // Total duration of the audio in WAV format, unit: s\n }\n ]\n },\n \"final_unit_deduction\": \"string\", // The deduction units of task\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit: ms\n \"updated_at\": 1722769557708 // Task update time, Unix timestamp, unit: ms\n }\n ]\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nQuery Parameters\npageNum\nint\nOptional\nDefault to 1\n\nPage number\n\nValue range: [1, 1000]\npageSize\nint\nOptional\nDefault to 30\n\nNumber of items per page\n\nValue range: [1, 500]\nPrevious chapter：Avatar\nNext chapter：Video to Audio\nCreate Task\nQuery Task (Single)\nQuery Task (List)\nThe Kling 3.0 Series Models API is Now Fully Available\n– All in One, One for All！\n\nModels Available in This Release\n\nKling 3.0 Motion Control, Kling Video 3.0, Kling Video 3.0 Omni, Kling Image 3.0, Kling Image 3.0 Omni\n\nRefer to \n\nKey Highlights of the Models\n\n3.0 All-in-One: A unified model for multi-modal input and output.\n\nMost powerful consistency across the universe: Subject consistency (supports cameo, subject with voice control, i2v + subject) and text consistency.\nNarrative control at your fingertips: More freedom, precision, and control—up to 15 seconds long, video scene cuts, ultra-high-definition storyboards/images, custom seconds.\nUpgraded native audio-visual output: Supports multiple speakers and languages (with accents).\n\nKling 3.0 Motion Control\n\nConsistent Facial Identity from any angle\nComplex Emotions faithfully reproduced\nHigh fidelity Restoration, Even with Face Occlusions\nConsistent Facial Clarity Across Dynamic Framing\n\nUser Guide ->\n\nKling Video 3.0\n\nCompared to 2.6, expected improvements:\n\nSupports subject upload in I2V scenarios for enhanced consistency\nSignificant improvement in multi-character referencing, especially for three-person scenarios\nSupports Japanese, Korean, and Spanish in addition to Chinese and English\nCapable of generating certain dialects and accents\nBetter distinction and control over different types of audio (speech, sound effects, BGM)\nImproved text retention in I2V scenarios\nSupports scene transitions, with up to 6 shots and customizable storyboarding\n\nUser Guide ->\n\nKling Video 3.0 Omni\n\nCompared to O1, expected improvements:\n\nNative audio-visual synchronization\nSupports video subject creation\nFurther improved consistency in reference-based tasks, especially for characters and products\nCombined capabilities of reference + storyboarding + audio-visual sync significantly enhance usability\nSupports scene transitions, with up to 6 shots\nExtended generation duration up to 15 seconds\n\nUser Guide ->\n\nKling Image 3.0\n\nHighly consistent feature retention\nPrecise response to detail modifications\nAccurate control over style and tone\nRich imaginative capabilities\n\nUser Guide ->\n\nKling Image 3.0 Omni\n\nEnhanced narrative sense\nNew storyboard image set generation, retaining reference image features with scene relevance\nDirect output of 2K/4K ultra-high-definition images\nFurther improved detail consistency\n\nUser Guide ->\n\nThank you for your support and understanding!\n\nI Got It", "Video to Audio": "Motion Control\nMulti-elements to video\nExtend Video\nLip Sync\nAvatar\nText to Audio\nVideo to Audio\nText to Speech\nVoice Clone\nImage Recognize\nElement\nEffects\nEffect Templates\nNEW\nVideo Effects\nImage Generation\nModels\nImage Omni\nImage Generation\nReference to Image\nExtend Image\nAI Multi-Shot\nVirtual Try-On\nOthers\nQuery user info\nPricing\nBilling Info\nPrepaid Resource Packs\nProtocols\nPrivacy Policy of API Service\nTerms of API Service\nAPI Service Level Agreement\nVideo to Audio\nCreate Task\nPOST\n/v1/audio/video-to-audio\ncURL\nCopy\nCollapse\ncurl --request POST \\\n --url https://api-singapore.klingai.com/v1/audio/video-to-audio \\\n --header 'Authorization: Bearer ' \\\n --header 'Content-Type: application/json' \\\n --data '{\n \"video_url\": \"https://p1-kling.klingai.com/kcdn/cdn-kcdn112452/kling-qa-test/20fps-7s.mov\",\n \"external_task_id\": \"\",\n \"callback_url\": \"\"\n }'\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes; Specific definitions can be found in \"Error Code\"\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system, is used to track requests and troubleshoot problems\n \"data\": {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_info\": { // Task creation parameters\n \"external_task_id\": \"string\" // Customer-defined task ID\n },\n \"task_status\": \"string\", // Task status, Enum values: submitted, processing, succeed, failed\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit: ms\n \"updated_at\": 1722769557708 // Task update time, Unix timestamp, unit: ms\n }\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nRequest Body\nvideo_id\nstring\nOptional\n\nThe ID of the video generated by the Kling AI\n\nEither the video_id parameter or the video_url parameter, cannot be empty or have a value at the same time.\nOnly supports videos generated within 30 days and with a duration between 3.0s and 20.0s.\nvideo_url\nstring\nOptional\n\nLink for uploaded video\n\nEither the video_id parameter or the video_url parameter, cannot be empty or have a value at the same time.\nOnly .mp4/.mov formats are supported. File size does not exceed 100MB. Video duration between 3.0s and 20.0s.\nsound_effect_prompt\nstring\nOptional\n\nSound effect prompt\n\nCannot exceed 200 characters\nbgm_prompt\nstring\nOptional\n\nBGM prompt\n\nCannot exceed 200 characters\nasmr_mode\nboolean\nOptional\nDefault to false\n\nEnable ASMR mode; This mode enhances detailed sound effects and is suitable for highly immersive content scenarios\n\ntrue means enabled, false means disabled (default)\nexternal_task_id\nstring\nOptional\n\nCustomized Task ID\n\nUsers can provide a customized task ID, which will not overwrite the system-generated task ID but can be used for task queries.\nPlease note that the customized task ID must be unique within a single user account.\ncallback_url\nstring\nOptional\n\nThe callback notification address for the result of this task. If configured, the server will actively notify when the task status changes\n\nThe specific message schema of the notification can be found in Callback Protocol\nQuery Task (Single)\nGET\n/v1/audio/video-to-audio/{id}\ncURL\nCopy\nCollapse\ncurl --request GET \\\n --url https://api-singapore.klingai.com/v1/audio/video-to-audio/{task_id} \\\n --header 'Authorization: Bearer '\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes; Specific definitions can be found in \"Error Code\"\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system, is used to track requests and troubleshoot problems\n \"data\": {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_status\": \"string\", // Task status, Enum values: submitted, processing, succeed, failed\n \"task_status_msg\": \"string\", // Task status information, displaying the failure reason when the task fails (such as triggering the content risk control of the platform, etc.)\n \"task_info\": { // Task creation parameters\n \"external_task_id\": \"string\", // Customer-defined task ID\n \"parent_video\": { // Original video information\n \"id\": \"string\", // Original video ID\n \"url\": \"string\", // Original video URL\n \"duration\": \"string\" // Original video duration, unit: s (seconds)\n }\n },\n \"task_result\": {\n \"videos\": [\n {\n \"id\": \"string\", // Generated video ID; globally unique\n \"url\": \"string\", // URL for generating videos (To ensure information security, generated images/videos will be cleared after 30 days. Please make sure to save them promptly.)\n \"duration\": \"string\" // Total video duration, unit: s (seconds)\n }\n ],\n \"audios\": [\n {\n \"id\": \"string\", // Generated audio ID; globally unique, will be cleared after 30 days\n \"url_mp3\": \"string\", // URL for generating audio in MP3 format (To ensure information security, generated audio will be cleared after 30 days. Please make sure to save them promptly.)\n \"url_wav\": \"string\", // URL for generating audio in WAV format (To ensure information security, generated audio will be cleared after 30 days. Please make sure to save them promptly.)\n \"duration_mp3\": \"string\", // Total MP3 audio duration, unit: s (seconds)\n \"duration_wav\": \"string\" // Total WAV audio duration, unit: s (seconds)\n }\n ]\n },\n \"final_unit_deduction\": \"string\", // The deduction units of task\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit: ms\n \"updated_at\": 1722769557708 // Task update time, Unix timestamp, unit: ms\n }\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nPath Parameters\ntask_id\nstring\nOptional\n\nThe task ID for audio generation\n\nRequest path parameter, fill the value directly in the request path\nYou can choose to query by external_task_id or task_id\nexternal_task_id\nstring\nOptional\n\nCustomized Task ID for audio generation\n\nThe external_task_id filled in when creating the task. You can choose to query by external_task_id or task_id\nQuery Task (List)\nGET\n/v1/audio/video-to-audio\ncURL\nCopy\nCollapse\ncurl --request GET \\\n --url 'https://api-singapore.klingai.com/v1/audio/video-to-audio?pageNum=1&pageSize=30' \\\n --header 'Authorization: Bearer '\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes; Specific definitions can be found in Error codes\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system, to track requests and troubleshoot problems\n \"data\": [\n {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_status\": \"string\", // Task status, Enum values: submitted, processing, succeed, failed\n \"task_status_msg\": \"string\", // Task status information, displaying the failure reason when the task fails (such as triggering the content risk control of the platform, etc.)\n \"task_info\": { // Task creation parameters\n \"external_task_id\": \"string\", // Customer-defined task ID\n \"parent_video\": { // Original video information\n \"id\": \"string\", // Original video ID\n \"url\": \"string\", // Original video URL\n \"duration\": \"string\" // Original video duration, unit: s (seconds)\n }\n },\n \"task_result\": {\n \"videos\": [\n {\n \"id\": \"string\", // Generated video ID; globally unique\n \"url\": \"string\", // URL for generating videos (To ensure information security, generated images/videos will be cleared after 30 days. Please make sure to save them promptly.)\n \"duration\": \"string\" // Total video duration, unit: s (seconds)\n }\n ],\n \"audios\": [\n {\n \"id\": \"string\", // Generated audio ID; globally unique, will be cleared after 30 days\n \"url_mp3\": \"string\", // URL for generating audio in MP3 format (To ensure information security, generated audio will be cleared after 30 days. Please make sure to save them promptly.)\n \"url_wav\": \"string\", // URL for generating audio in WAV format (To ensure information security, generated audio will be cleared after 30 days. Please make sure to save them promptly.)\n \"duration_mp3\": \"string\", // Total MP3 audio duration, unit: s (seconds)\n \"duration_wav\": \"string\" // Total WAV audio duration, unit: s (seconds)\n }\n ]\n },\n \"final_unit_deduction\": \"string\", // The deduction units of task\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit: ms\n \"updated_at\": 1722769557708 // Task update time, Unix timestamp, unit: ms\n }\n ]\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nQuery Parameters\npageNum\nint\nOptional\nDefault to 1\n\nPage number\n\nValue range: [1, 1000]\npageSize\nint\nOptional\nDefault to 30\n\nNumber of items per page\n\nValue range: [1, 500]\nPrevious chapter：Text to Audio\nNext chapter：Text to Speech\nCreate Task\nQuery Task (Single)\nQuery Task (List)\nThe Kling 3.0 Series Models API is Now Fully Available\n– All in One, One for All！\n\nModels Available in This Release\n\nKling 3.0 Motion Control, Kling Video 3.0, Kling Video 3.0 Omni, Kling Image 3.0, Kling Image 3.0 Omni\n\nRefer to \n\nKey Highlights of the Models\n\n3.0 All-in-One: A unified model for multi-modal input and output.\n\nMost powerful consistency across the universe: Subject consistency (supports cameo, subject with voice control, i2v + subject) and text consistency.\nNarrative control at your fingertips: More freedom, precision, and control—up to 15 seconds long, video scene cuts, ultra-high-definition storyboards/images, custom seconds.\nUpgraded native audio-visual output: Supports multiple speakers and languages (with accents).\n\nKling 3.0 Motion Control\n\nConsistent Facial Identity from any angle\nComplex Emotions faithfully reproduced\nHigh fidelity Restoration, Even with Face Occlusions\nConsistent Facial Clarity Across Dynamic Framing\n\nUser Guide ->\n\nKling Video 3.0\n\nCompared to 2.6, expected improvements:\n\nSupports subject upload in I2V scenarios for enhanced consistency\nSignificant improvement in multi-character referencing, especially for three-person scenarios\nSupports Japanese, Korean, and Spanish in addition to Chinese and English\nCapable of generating certain dialects and accents\nBetter distinction and control over different types of audio (speech, sound effects, BGM)\nImproved text retention in I2V scenarios\nSupports scene transitions, with up to 6 shots and customizable storyboarding\n\nUser Guide ->\n\nKling Video 3.0 Omni\n\nCompared to O1, expected improvements:\n\nNative audio-visual synchronization\nSupports video subject creation\nFurther improved consistency in reference-based tasks, especially for characters and products\nCombined capabilities of reference + storyboarding + audio-visual sync significantly enhance usability\nSupports scene transitions, with up to 6 shots\nExtended generation duration up to 15 seconds\n\nUser Guide ->\n\nKling Image 3.0\n\nHighly consistent feature retention\nPrecise response to detail modifications\nAccurate control over style and tone\nRich imaginative capabilities\n\nUser Guide ->\n\nKling Image 3.0 Omni\n\nEnhanced narrative sense\nNew storyboard image set generation, retaining reference image features with scene relevance\nDirect output of 2K/4K ultra-high-definition images\nFurther improved detail consistency\n\nUser Guide ->\n\nThank you for your support and understanding!\n\nI Got It", "Text to Speech": "Motion Control\nMulti-elements to video\nExtend Video\nLip Sync\nAvatar\nText to Audio\nVideo to Audio\nText to Speech\nVoice Clone\nImage Recognize\nElement\nEffects\nEffect Templates\nNEW\nVideo Effects\nImage Generation\nModels\nImage Omni\nImage Generation\nReference to Image\nExtend Image\nAI Multi-Shot\nVirtual Try-On\nOthers\nQuery user info\nPricing\nBilling Info\nPrepaid Resource Packs\nProtocols\nPrivacy Policy of API Service\nTerms of API Service\nAPI Service Level Agreement\nTTS\nCreate Task\nPOST\n/v1/audio/tts\ncURL\nCopy\nCollapse\ncurl --request POST \\\n --url https://api-singapore.klingai.com/v1/audio/tts \\\n --header 'Authorization: Bearer ' \\\n --header 'Content-Type: application/json' \\\n --data '{\n \"text\": \"Throughout my time in college, several memorable event left a significant impact on my life\",\n \"voice_id\": \"oversea_male1\",\n \"voice_language\": \"en\",\n \"voice_speed\": 1\n }'\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes; Specific definitions can be found in Error codes\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system, is used to track requests and troubleshoot problems\n \"data\": {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_status\": \"string\", // Task status, Enum values: submitted, processing, succeed, failed\n \"task_status_msg\": \"string\", // Task status information, displaying the failure reason when the task fails (such as triggering the content risk control of the platform, etc.)\n \"task_result\": {\n \"audios\": [\n {\n \"id\": \"string\", // Generated sound ID; globally unique, will be cleared after 30 days\n \"url\": \"string\", // URL for generating sounds，such as https://p1.a.kwimgs.com/bs2/upload-ylab-stunt/special-effect/output/HB1_PROD_ai_web_46554461/-2878350957757294165/output.mp3(To ensure information security, generated images/videos will be cleared after 30 days. Please make sure to save them promptly.)\n \"duration\": \"string\" // Total audio duration, unit: s (seconds)\n }\n ]\n },\n \"final_unit_deduction\": \"string\", // The deduction units of task\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit: ms\n \"updated_at\": 1722769557708 // Task update time, Unix timestamp, unit: ms\n }\n}\n\nText-to-Speech synthesis API for generating audio from text.\n\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nRequest Body\ntext\nstring\nRequired\n\nText Content for Audio Synthesis\n\nThe maximum length of the text content is 1000 characters; content that is too long will return an error code and other information.\nThe system will validate the text content; if there are issues, it will return an error code and other information.\nvoice_id\nstring\nRequired\n\nVoice ID\n\nThe system offers a variety of voice options to choose from. For specific voice effects, voice IDs, and corresponding voice languages, see Voice Guide. Voice previews do not support custom scripts.\nVoice preview file naming convention: Voice Name#Voice ID#Voice Language\nvoice_language\nstring\nRequired\nDefault to zh\n\nVoice Language\n\nEnum values：\nzh\nen\nThe voice language corresponds to the Voice ID, as detailed above.\nvoice_speed\nfloat\nOptional\nDefault to 1.0\n\nSpeech Rate\n\nValid range: [0.8, 2.0], accurate to one decimal place; values outside this range will be automatically rounded.\nPrevious chapter：Video to Audio\nNext chapter：Voice Clone\nThe Kling 3.0 Series Models API is Now Fully Available\n– All in One, One for All！\n\nModels Available in This Release\n\nKling 3.0 Motion Control, Kling Video 3.0, Kling Video 3.0 Omni, Kling Image 3.0, Kling Image 3.0 Omni\n\nRefer to \n\nKey Highlights of the Models\n\n3.0 All-in-One: A unified model for multi-modal input and output.\n\nMost powerful consistency across the universe: Subject consistency (supports cameo, subject with voice control, i2v + subject) and text consistency.\nNarrative control at your fingertips: More freedom, precision, and control—up to 15 seconds long, video scene cuts, ultra-high-definition storyboards/images, custom seconds.\nUpgraded native audio-visual output: Supports multiple speakers and languages (with accents).\n\nKling 3.0 Motion Control\n\nConsistent Facial Identity from any angle\nComplex Emotions faithfully reproduced\nHigh fidelity Restoration, Even with Face Occlusions\nConsistent Facial Clarity Across Dynamic Framing\n\nUser Guide ->\n\nKling Video 3.0\n\nCompared to 2.6, expected improvements:\n\nSupports subject upload in I2V scenarios for enhanced consistency\nSignificant improvement in multi-character referencing, especially for three-person scenarios\nSupports Japanese, Korean, and Spanish in addition to Chinese and English\nCapable of generating certain dialects and accents\nBetter distinction and control over different types of audio (speech, sound effects, BGM)\nImproved text retention in I2V scenarios\nSupports scene transitions, with up to 6 shots and customizable storyboarding\n\nUser Guide ->\n\nKling Video 3.0 Omni\n\nCompared to O1, expected improvements:\n\nNative audio-visual synchronization\nSupports video subject creation\nFurther improved consistency in reference-based tasks, especially for characters and products\nCombined capabilities of reference + storyboarding + audio-visual sync significantly enhance usability\nSupports scene transitions, with up to 6 shots\nExtended generation duration up to 15 seconds\n\nUser Guide ->\n\nKling Image 3.0\n\nHighly consistent feature retention\nPrecise response to detail modifications\nAccurate control over style and tone\nRich imaginative capabilities\n\nUser Guide ->\n\nKling Image 3.0 Omni\n\nEnhanced narrative sense\nNew storyboard image set generation, retaining reference image features with scene relevance\nDirect output of 2K/4K ultra-high-definition images\nFurther improved detail consistency\n\nUser Guide ->\n\nThank you for your support and understanding!\n\nI Got It", "Voice Clone": "Motion Control\nMulti-elements to video\nExtend Video\nLip Sync\nAvatar\nText to Audio\nVideo to Audio\nText to Speech\nVoice Clone\nImage Recognize\nElement\nEffects\nEffect Templates\nNEW\nVideo Effects\nImage Generation\nModels\nImage Omni\nImage Generation\nReference to Image\nExtend Image\nAI Multi-Shot\nVirtual Try-On\nOthers\nQuery user info\nPricing\nBilling Info\nPrepaid Resource Packs\nProtocols\nPrivacy Policy of API Service\nTerms of API Service\nAPI Service Level Agreement\nCustom Voice\nCreate Custom Voice\nPOST\n/v1/general/custom-voices\ncURL\nCopy\nCollapse\ncurl --request POST \\\n --url https://api-singapore.klingai.com/v1/general/custom-voices \\\n --header 'Authorization: Bearer ' \\\n --header 'Content-Type: application/json' \\\n --data '{\n \"video_id\": \"\",\n \"voice_url\": \"https://p2-kling.klingai.com/kcdn/cdn-kcdn112452/kling-qa-test/out.mp3\",\n \"voice_name\": \"Custom Voice\",\n \"callback_url\": \"\"\n }'\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error Codes；Specific definitions can be found in Error codes\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system, is used to track requests and troubleshoot problems\n \"data\": {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_info\": { // Task creation parameters\n \"external_task_id\": \"string\" // Customer-defined task ID\n },\n \"task_status\": \"string\", // Task status, Enum values：submitted、processing、succeed、failed\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit ms\n \"updated_at\": 1722769557708 // Task update time, Unix timestamp, unit ms\n }\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nRequest Body\nvoice_name\nstring\nRequired\n\nVoice Name\n\nThe maximum length of text content is 20 characters\nThe voices that are no longer used after creation can be deleted through the API\nvoice_url\nstring\nOptional\n\nThe URL of voice data file\n\nSupports .mp3 / .wav audio file and .mp4 / .mov video file\nThe voice needs to be clean and free of noise, with only one type of human voice present, with a duration of no less than 5 seconds and no longer than 30 seconds\nvideo_id\nstring\nOptional\n\nGenerated video ID, which can provide audio materials by referencing historical works\n\nOnly videos that meet the following conditions can be used to customize voice:\nThe video is generated on V2.6 model and the value of sound parameter is on\nThe video is generated through Avatar API\nThe video is generated through Lip-Sync API\nThe voice needs to be clean and free of noise, with only one type of human voice present, with a duration of no less than 5 seconds and no longer than 30 seconds\ncallback_url\nstring\nOptional\n\nThe callback notification address for the result of this task. If configured, the server will actively notify when the task status changes.\n\nFor specific message schema, see Callback Protocol\nexternal_task_id\nstring\nOptional\n\nCustomized Task ID\n\nWill not overwrite system-generated task ID, but supports querying task by this ID\nPlease note that the customized task ID must be unique within a single user account.\nQuery Custom Voice (Single)\nGET\n/v1/general/custom-voices/{id}\ncURL\nCopy\nCollapse\ncurl --request GET \\\n --url 'https://api-singapore.klingai.com/v1/general/custom-voices/{id}' \\\n --header 'Authorization: Bearer '\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes；Specific definitions can be found in \"Error Code\"\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system, is used to track requests and troubleshoot problems\n \"data\": {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_status\": \"string\", // Task status, Enum values：submitted、processing、succeed、failed\n \"task_status_msg\": \"string\", // Task status information, displaying the failure reason when the task fails (such as triggering the content risk control of the platform, etc.)\n \"task_info\": { // Task creation parameters\n \"external_task_id\": \"string\" // Customer-defined task ID\n },\n \"task_result\": {\n \"voices\": [\n {\n \"voice_id\": \"string\", // Generated voice ID; globally unique\n \"voice_name\": \"string\", // Generated voice name\n \"trial_url\": \"string\", // URL for generating videos\n \"owned_by\": \"kling\" // Voice source, kling is the official voice library, and others are the creator's ID\n }\n ]\n },\n \"final_unit_deduction\": \"string\", // The deduction units of task\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit: ms\n \"updated_at\": 1722769557708 // Task update time, Unix timestamp, unit: ms\n }\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nPath Parameters\ntask_id\nstring\nRequired\n\nThe task ID of the element creation task. Request path parameter, directly fill the value in the request path.\n\nexternal_task_id\nstring\nOptional\n\nCustomized Task ID for audio generation\n\nThe external_task_id filled in when creating the task. You can choose to query by external_task_id or task_id\nWhen creating a task, you can choose to query by external_task_id or task_id.\nQuery Custom Voice (List)\nGET\n/v1/general/custom-voices\ncURL\nCopy\nCollapse\ncurl --request GET \\\n --url 'https://api-singapore.klingai.com/v1/general/custom-voices?pageNum=1&pageSize=30' \\\n --header 'Authorization: Bearer '\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes；Specific definitions can be found in \"Error Code\"\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system, is used to track requests and troubleshoot problems\n \"data\": [\n {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_status\": \"string\", // Task status, Enum values：submitted、processing、succeed、failed\n \"task_status_msg\": \"string\", // Task status information, displaying the failure reason when the task fails (such as triggering the content risk control of the platform, etc.)\n \"task_info\": { // Task creation parameters\n \"external_task_id\": \"string\" // Customer-defined task ID\n },\n \"task_result\": {\n \"voices\": [\n {\n \"voice_id\": \"string\", // Generated voice ID; globally unique\n \"voice_name\": \"string\", // Generated voice name\n \"trial_url\": \"string\", // URL for generating videos\n \"owned_by\": \"kling\" // Voice source, kling is the official voice library, and others are the creator's ID\n }\n ]\n }\n \"final_unit_deduction\": \"string\", // The deduction units of task\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit: ms\n \"updated_at\": 1722769557708, // Task update time, Unix timestamp, unit: ms\n }\n ]\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nQuery Parameters\npageNum\nint\nOptional\nDefault to 1\n\nPage number\n\nValue range: [1, 1000]\n\npageSize\nint\nOptional\nDefault to 30\n\nData volume per page\n\nValue range: [1, 1000]\n\nQuery Presets Voice (List)\nGET\n/v1/general/presets-voices\ncURL\nCopy\nCollapse\ncurl --request GET \\\n --url 'https://api-singapore.klingai.com/v1/general/presets-voices?pageNum=1&pageSize=30' \\\n --header 'Authorization: Bearer '\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes；Specific definitions can be found in \"Error Code\"\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system, is used to track requests and troubleshoot problems\n \"data\": [\n {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_status\": \"string\", // Task status, Enum values：submitted、processing、succeed、failed\n \"task_status_msg\": \"string\", // Task status information, displaying the failure reason when the task fails (such as triggering the content risk control of the platform, etc.)\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit: ms\n \"updated_at\": 1722769557708, // Task update time, Unix timestamp, unit: ms\n \"task_result\": {\n \"voices\": [\n {\n \"voice_id\": \"string\", // Official voice ID; globally unique\n \"voice_name\": \"string\", // Official voice name\n \"trial_url\": \"string\", // URL for official videos\n \"owned_by\": \"kling\" // Voice source, kling is the official voice library, and others are the creator's ID\n }\n ]\n }\n }\n ]\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nQuery Parameters\npageNum\nint\nOptional\nDefault to 1\n\nPage number\n\nValue range: [1, 1000]\n\npageSize\nint\nOptional\nDefault to 30\n\nData volume per page\n\nValue range: [1, 1000]\n\nDelete Custom Voice\nPOST\n/v1/general/delete-voices\ncURL\nCopy\nCollapse\ncurl --request POST \\\n --url https://api-singapore.klingai.com/v1/general/delete-voices \\\n --header 'Authorization: Bearer ' \\\n --header 'Content-Type: application/json' \\\n --data '{\n \"voice_id\": \"850087542757535834\"\n }'\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes；Specific definitions can be found in \"Error Code\"\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system\n \"data\": {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_status\": \"string\" // Task status, Enum values：submitted、processing、succeed、failed\n }\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nRequest Body\nvoice_id\nstring\nRequired\n\nThe ID of the voice to be deleted, only supports deleting custom voices\n\nPrevious chapter：Text to Speech\nNext chapter：Image Recognize\nCreate Custom Voice\nQuery Custom Voice (Single)\nQuery Custom Voice (List)\nQuery Presets Voice (List)\nDelete Custom Voice\nThe Kling 3.0 Series Models API is Now Fully Available\n– All in One, One for All！\n\nModels Available in This Release\n\nKling 3.0 Motion Control, Kling Video 3.0, Kling Video 3.0 Omni, Kling Image 3.0, Kling Image 3.0 Omni\n\nRefer to \n\nKey Highlights of the Models\n\n3.0 All-in-One: A unified model for multi-modal input and output.\n\nMost powerful consistency across the universe: Subject consistency (supports cameo, subject with voice control, i2v + subject) and text consistency.\nNarrative control at your fingertips: More freedom, precision, and control—up to 15 seconds long, video scene cuts, ultra-high-definition storyboards/images, custom seconds.\nUpgraded native audio-visual output: Supports multiple speakers and languages (with accents).\n\nKling 3.0 Motion Control\n\nConsistent Facial Identity from any angle\nComplex Emotions faithfully reproduced\nHigh fidelity Restoration, Even with Face Occlusions\nConsistent Facial Clarity Across Dynamic Framing\n\nUser Guide ->\n\nKling Video 3.0\n\nCompared to 2.6, expected improvements:\n\nSupports subject upload in I2V scenarios for enhanced consistency\nSignificant improvement in multi-character referencing, especially for three-person scenarios\nSupports Japanese, Korean, and Spanish in addition to Chinese and English\nCapable of generating certain dialects and accents\nBetter distinction and control over different types of audio (speech, sound effects, BGM)\nImproved text retention in I2V scenarios\nSupports scene transitions, with up to 6 shots and customizable storyboarding\n\nUser Guide ->\n\nKling Video 3.0 Omni\n\nCompared to O1, expected improvements:\n\nNative audio-visual synchronization\nSupports video subject cr", "Image Recognize": "Motion Control\nMulti-elements to video\nExtend Video\nLip Sync\nAvatar\nText to Audio\nVideo to Audio\nText to Speech\nVoice Clone\nImage Recognize\nElement\nEffects\nEffect Templates\nNEW\nVideo Effects\nImage Generation\nModels\nImage Omni\nImage Generation\nReference to Image\nExtend Image\nAI Multi-Shot\nVirtual Try-On\nOthers\nQuery user info\nPricing\nBilling Info\nPrepaid Resource Packs\nProtocols\nPrivacy Policy of API Service\nTerms of API Service\nAPI Service Level Agreement\nImage Recognize\nImage Recognize\nPOST\n/v1/videos/image-recognize\ncURL\nCopy\nCollapse\ncurl --request POST \\\n --url https://api-singapore.klingai.com/v1/videos/image-recognize \\\n --header 'Authorization: Bearer ' \\\n --header 'Content-Type: application/json' \\\n --data '{\n \"image\": \"https://p2-kling.klingai.com/kcdn/cdn-kcdn112452/kling-qa-test/multi-1.png\"\n }'\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes; Specific definitions can be found in Error codes\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system, is used to track requests and troubleshoot problems\n \"data\": {\n \"final_unit_deduction\": \"string\", // The deduction units of task\n \"task_result\": {\n \"images\": [\n {\n \"type\": \"object_seg\", // Identification of subject recognition results\n \"is_contain\": true, // Has the subject been identified; Boolean value\n \"url\": \"string\" //URL for generating videos，such as https://p1.a.kwimgs.com/bs2/upload-ylab-stunt/special-effect/output/HB1_PROD_ai_web_46554461/-2878350957757294165/output.png（请注意，为保障信息安全，生成的图片/视频会在30天后被清理，请及时转存）\n },\n {\n \"type\": \"head_seg\", // Identification of facial recognition results for individuals with hair included\n \"is_contain\": true, // Has the subject been identified; Boolean value\n \"url\": \"string\" //URL for generating videos，such as https://p1.a.kwimgs.com/bs2/upload-ylab-stunt/special-effect/output/HB1_PROD_ai_web_46554461/-2878350957757294165/output.png（请注意，为保障信息安全，生成的图片/视频会在30天后被清理，请及时转存）\n },\n {\n \"type\": \"face_seg\", // Identification of facial recognition results for individuals without hair included\n \"is_contain\": true, // Has the subject been identified; Boolean value\n \"url\": \"string\" //URL for generating videos，such as https://p1.a.kwimgs.com/bs2/upload-ylab-stunt/special-effect/output/HB1_PROD_ai_web_46554461/-2878350957757294165/output.png（请注意，为保障信息安全，生成的图片/视频会在30天后被清理，请及时转存）\n },\n {\n \"type\": \"cloth_seg\", // Identification of clothing recognition results\n \"is_contain\": true, // Has the subject been identified; Boolean value\n \"url\": \"string\" //URL for generating videos，such as https://p1.a.kwimgs.com/bs2/upload-ylab-stunt/special-effect/output/HB1_PROD_ai_web_46554461/-2878350957757294165/output.png（请注意，为保障信息安全，生成的图片/视频会在30天后被清理，请及时转存）\n }\n ]\n },\n \"final_unit_deduction\": \"string\" // The deduction units of task\n }\n}\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nRequest Body\nimage\nstring\nRequired\n\nImage to be recognized\n\nSupport inputting image Base64 encoding or image URL (ensure accessibility).\n\nPlease note, if you use the Base64 method, make sure all image data parameters you pass are in Base64 encoding format. When submitting data, do not add any prefixes to the Base64-encoded string, such as data:image/png;base64,. The correct parameter format should be the Base64-encoded string itself. Please provide only the Base64-encoded string portion so that the system can correctly process and parse your data.\n\nSupported image formats: .jpg / .jpeg / .png. The image file size cannot exceed 10MB, and the width and height dimensions of the image shall not be less than 300px, and the aspect ratio of the image should be between 1:2.5 ~ 2.5:1.\n\nPrevious chapter：Voice Clone\nNext chapter：Element\nThe Kling 3.0 Series Models API is Now Fully Available\n– All in One, One for All！\n\nModels Available in This Release\n\nKling 3.0 Motion Control, Kling Video 3.0, Kling Video 3.0 Omni, Kling Image 3.0, Kling Image 3.0 Omni\n\nRefer to \n\nKey Highlights of the Models\n\n3.0 All-in-One: A unified model for multi-modal input and output.\n\nMost powerful consistency across the universe: Subject consistency (supports cameo, subject with voice control, i2v + subject) and text consistency.\nNarrative control at your fingertips: More freedom, precision, and control—up to 15 seconds long, video scene cuts, ultra-high-definition storyboards/images, custom seconds.\nUpgraded native audio-visual output: Supports multiple speakers and languages (with accents).\n\nKling 3.0 Motion Control\n\nConsistent Facial Identity from any angle\nComplex Emotions faithfully reproduced\nHigh fidelity Restoration, Even with Face Occlusions\nConsistent Facial Clarity Across Dynamic Framing\n\nUser Guide ->\n\nKling Video 3.0\n\nCompared to 2.6, expected improvements:\n\nSupports subject upload in I2V scenarios for enhanced consistency\nSignificant improvement in multi-character referencing, especially for three-person scenarios\nSupports Japanese, Korean, and Spanish in addition to Chinese and English\nCapable of generating certain dialects and accents\nBetter distinction and control over different types of audio (speech, sound effects, BGM)\nImproved text retention in I2V scenarios\nSupports scene transitions, with up to 6 shots and customizable storyboarding\n\nUser Guide ->\n\nKling Video 3.0 Omni\n\nCompared to O1, expected improvements:\n\nNative audio-visual synchronization\nSupports video subject creation\nFurther improved consistency in reference-based tasks, especially for characters and products\nCombined capabilities of reference + storyboarding + audio-visual sync significantly enhance usability\nSupports scene transitions, with up to 6 shots\nExtended generation duration up to 15 seconds\n\nUser Guide ->\n\nKling Image 3.0\n\nHighly consistent feature retention\nPrecise response to detail modifications\nAccurate control over style and tone\nRich imaginative capabilities\n\nUser Guide ->\n\nKling Image 3.0 Omni\n\nEnhanced narrative sense\nNew storyboard image set generation, retaining reference image features with scene relevance\nDirect output of 2K/4K ultra-high-definition images\nFurther improved detail consistency\n\nUser Guide ->\n\nThank you for your support and understanding!\n\nI Got It", "Element": "Motion Control\nMulti-elements to video\nExtend Video\nLip Sync\nAvatar\nText to Audio\nVideo to Audio\nText to Speech\nVoice Clone\nImage Recognize\nElement\nEffects\nEffect Templates\nNEW\nVideo Effects\nImage Generation\nModels\nImage Omni\nImage Generation\nReference to Image\nExtend Image\nAI Multi-Shot\nVirtual Try-On\nOthers\nQuery user info\nPricing\nBilling Info\nPrepaid Resource Packs\nProtocols\nPrivacy Policy of API Service\nTerms of API Service\nAPI Service Level Agreement\nElement\nCreate Element\nPOST\n/v1/general/advanced-custom-elements\n200\nCopy\nCollapse\n{\n \"code\": 0, //Error codes; specific definitions see Error codes\n \"message\": \"string\", //Error information\n \"request_id\": \"string\", //Request ID, generated by the system, for tracking and troubleshooting\n \"data\": {\n \"task_id\": \"string\", //Task ID, generated by the system\n \"task_info\": { //Task creation parameters\n \"external_task_id\": \"string\" //Customer-defined task ID\n },\n \"task_status\": \"string\", //Task status: submitted, processing, succeed, failed\n \"created_at\": 1722769557708, //Task creation time, Unix timestamp, ms\n \"updated_at\": 1722769557708 //Task update time, Unix timestamp, ms\n }\n}\n💡\n\nThe service related to creating entities has been upgraded to a brand new version. If you need to browse the old version, please proceed to:Kling AI (OLD VERSION) ELEMENTS API Specification\n\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nRequest Body\nelement_name\nstring\nRequired\n\nElement Name\n\nMust not exceed 20 characters.\n\nelement_description\nstring\nRequired\n\nElement Description\n\nMust not exceed 100 characters.\n\nreference_type\nstring\nRequired\n\nReference Method\n\nEnum values：\nvideo_refer\nimage_refer\n\nvideo_refer: Video Character Elements, at this time, the subject's appearance will be defined with reference to element_video_list.\n\nimage_refer: Multi-Image Elements, whose appearance will be defined with reference to the element_image_list.\n\nThe scope of availability differs between entities customized through videos and those customized through images. Please refer to the capability map and parameter specifications for details.\n\nelement_image_list\nobject\nOptional\nDefault to None\n\nThe main reference image allows for the setting of the element and its details through multiple images.\n\nInclude front reference images and reference images from other angles or close-ups: at least one frontal reference image (frontal_image), and 1 to 3 additional reference images (image_url) that differ from the front.\n\nLoad with key:value format as follows:\n\n\"element_image_list\": {\n\"frontal_image\": \"image_url_0\", \n\"refer_images\": [{ \"image_url\": \"image_url_1\" }, ...] \n}\n\nSupports inputting image Base64 encoding or image URL (ensure accessibility).\n\nSupported image formats: .jpg / .jpeg / .png. Image file size cannot exceed 10MB, dimensions not less than 300px, aspect ratio between 1:2.5 ~ 2.5:1.\n\nWhen reference_type is image_refer, this parameter is required.\n\nelement_video_list\nobject\nOptional\nDefault to None\n\nThe element is referenced by the video, and its details can be set through the video.\n\nAudio videos can be uploaded. If the audio video contains human voice, it will trigger voice customization (customization + inclusion in voice library + binding with the element).\n\nCurrently, only realistic-style humanoid figures can be customized through video.\n\nRequired when referencing videos; invalid when referencing images.\n\nStructure: element_video_list: { refer_videos: [{ video_url: \"video_url_1\" }] }. Only .mp4/.mov formats. Duration 3s–8s, 1080P, aspect ratio 16:9 or 9:16. At most 1 video, size not exceeding 200MB. video_url must not be empty.\n\n\"element_video_list\": {\n\"refer_videos\": [{ \"video_url\": \"video_url_1\" }, ...] \n}\n\nVideo-customized elements are only supported for kling-video-o3 and later models.\n\nelement_voice_id\nstring\nOptional\nDefault to None\n\nThe voice ID of element can be bound to existing tone colors in the tone library\n\nWhen the current parameter is empty, the current entity is not bound to a tone color.\nThe ID can be obtained through the voice-related API. For details, see Voice Guide\ntag_list\narray\nOptional\nDefault to None\n\nConfigure tags for the subject, one subject can configure multiple tags.\n\nStructure: tag_list: [{ tag_id: \"o_101\" }, { tag_id: \"o_102\" }, ...]. Tag ID and name: o_101 Hottest, o_102 Character, o_103 Animal, o_104 Item, o_105 Costume, o_106 Scene, o_107 Effect, o_108 Others.\n\n\"tag_list\": [ { \"tag_id\": \"o_101\" }, { \"tag_id\": \"o_102\" } ]\n\nTag and tag_id correspondence:\n\ntag_id\ttag_name\no_101\tHottest\no_102\tCharacter\no_103\tAnimal\no_104\tItem\no_105\tCostume\no_106\tScene\no_107\tEffect\no_108\tOthers\ncallback_url\nstring\nOptional\n\nThe callback notification address for the result of this task. If configured, the server will actively notify when the task status changes.\n\nFor the specific message schema, see Callback Protocol\nexternal_task_id\nstring\nOptional\n\nCustomized Task ID. Users can provide a customized task ID, which will not overwrite the system-generated task ID but can be used for task queries. Please note that the customized task ID must be unique within a single user account.\n\nInvocation examples\nCreate Multi-Image Elements\ncurl --location 'https://xxx/v1/general/advanced-custom-elements/' \\\n--header 'Authorization: Bearer xxx' \\\n--header 'Content-Type: application/json\\' \\\n--data '{\n \"element_name\": \"xxx\",\n \"element_description\": \"xxx\",\n \"reference_type\": \"image_refer\",\n \"element_image_list\": {\n \"frontal_image\": \"xxx\",\n \"refer_images\": [\n {\"image_url\": \"xxx\"},\n {\"image_url\": \"xxx\"}\n ]\n },\n \"element_voice_id\": string,\n \"callback_url\": \"xxx\",\n \"external_task_id\": \"\",\n \"tag_list\": [\n {\n \"tag_id\": \"xxx\"\n }\n ]\n }'\nCreate Video Character Elements\ncurl --location 'https://xxx/v1/general/advanced-custom-elements/' \\\n--header 'Authorization: Bearer xxx' \\\n--header 'Content-Type: application/json\\' \\\n--data '{\n \"element_name\": \"xxx\",\n \"element_description\": \"xxx\",\n \"reference_type\": \"video_refer\",\n \"element_video_list\": {\n \"refer_videos\": [\n {\n \"video_url\": \"xxx\"\n }\n ]\n },\n \"element_voice_id\": string,\n \"callback_url\": \"xxx\",\n \"external_task_id\": \"\",\n \"tag_list\": [\n {\n \"tag_id\": \"xxx\"\n }\n ]\n}'\nQuery Custom Element (Single)\nGET\n/v1/general/advanced-custom-elements/{id}\n200\nCopy\nCollapse\n{\n \"code\": 0, //Error codes; specific definitions see Error codes\n \"message\": \"string\", //Error information\n \"request_id\": \"string\", //Request ID, generated by the system\n \"data\": {\n \"task_id\": \"string\", //Task ID, generated by the system\n \"task_status\": \"string\", //Task status: submitted, processing, succeed, failed\n \"task_status_msg\": \"string\", //Task status message, failure reason when failed\n \"task_info\": { //Task creation parameters\n \"external_task_id\": \"string\" //Customer-defined task ID\n },\n \"task_result\": {\n \"elements\": [\n {\n \"element_id\": 0,\n \"element_name\": \"string\",\n \"element_description\": \"string\",\n \"reference_type\": \"video_refer\",\n \"element_image_list\": {},\n \"element_video_list\": {},\n \"element_voice_info\": {\n \"voice_id\": \"string\", //Custom voice ID; globally unique\n \"voice_name\": \"string\", //Custom voice name\n \"trial_url\": \"string\", //Trial audio download URL\n \"owned_by\": \"kling\" //Voice source, kling is official, number is creator ID\n },\n \"tag_list\": [],\n \"owned_by\": \"kling\", //Element source, kling is official element library\n \"status\": \"succeed\" //Element status: succeed when normal, deleted when removed\n }\n ]\n },\n \"final_unit_deduction\": \"string\", //Final unit deduction for the task\n \"created_at\": 1722769557708, //Task creation time, Unix timestamp, ms\n \"updated_at\": 1722769557708 //Task update time, Unix timestamp, ms\n }\n}\n💡\n\nThe service related to creating entities has been upgraded to a brand new version. If you need to browse the old version, please proceed to:Kling AI (OLD VERSION) ELEMENTS API Specification\n\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nPath Parameters\ntask_id\nstring\nRequired\n\nThe task ID of the element creation task. Request path parameter, directly fill the value in the request path.\n\nexternal_task_id\nstring\nOptional\n\nCustomized Task ID for audio generation\n\nThe external_task_id filled in when creating the task. You can choose to query by external_task_id or task_id\nWhen creating a task, you can choose to query by external_task_id or task_id.\nQuery Custom Element (List)\nGET\n/v1/general/advanced-custom-elements\n200\nCopy\nCollapse\n{\n \"code\": 0, //Error codes; specific definitions see Error codes\n \"message\": \"string\", //Error information\n \"request_id\": \"string\", //Request ID, generated by the system\n \"data\": [\n {\n \"task_id\": \"string\", //Task ID, generated by the system\n \"task_status\": \"string\", //Task status: submitted, processing, succeed, failed\n \"task_status_msg\": \"string\", //Task status message\n \"task_info\": { \"external_task_id\": \"string\" }, //Task creation parameters\n \"task_result\": {\n \"elements\": [\n {\n \"element_id\": 0,\n \"element_name\": \"string\",\n \"element_description\": \"string\",\n \"reference_type\": \"video_refer\",\n \"element_image_list\": {},\n \"element_video_list\": {},\n \"element_voice_info\": {},\n \"tag_list\": [],\n \"owned_by\": \"kling\", //Element source\n \"status\": \"succeed\" //Element status: succeed, deleted\n }\n ]\n },\n \"final_unit_deduction\": \"string\", //Final unit deduction\n \"created_at\": 1722769557708, //Task creation time, Unix timestamp, ms\n \"updated_at\": 1722769557708 //Task update time, Unix timestamp, ms\n }\n ]\n}\n💡\n\nThe service related to creating entities has been upgraded to a brand new version. If you need to browse the old version, please proceed to:Kling AI (OLD VERSION) ELEMENTS API Specification\n\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nQuery Parameters\npageNum\nint\nOptional\nDefault to 1\n\nPage number\n\nValue range: [1, 1000]\n\npageSize\nint\nOptional\nDefault to 30\n\nData volume per page\n\nValue range: [1, 500]\n\nQuery Presets Element (List)\nGET\n/v1/general/advanced-presets-elements\n200\nCopy\nCollapse\n{\n \"code\": 0, //Error codes; specific definitions see Error codes\n \"message\": \"string\", //Error information\n \"request_id\": \"string\", //Request ID, generated by the system\n \"data\": [\n {\n \"task_id\": \"string\", //Task ID, generated by the system\n \"task_status\": \"string\", //Task status: submitted, processing, succeed, failed\n \"task_status_msg\": \"string\", //Task status message\n \"task_info\": { \"external_task_id\": \"string\" }, //Task creation parameters\n \"task_result\": {\n \"elements\": [\n {\n \"element_id\": 0,\n \"element_name\": \"string\",\n \"element_description\": \"string\",\n \"reference_type\": \"video_refer\",\n \"element_image_list\": {},\n \"element_video_list\": {},\n \"element_voice_info\": {},\n \"tag_list\": [],\n \"owned_by\": \"kling\", //Element source\n \"status\": \"succeed\" //Element status: succeed, deleted\n }\n ]\n },\n \"final_unit_deduction\": \"string\", //Final unit deduction\n \"created_at\": 1722769557708, //Task creation time, Unix timestamp, ms\n \"", "Video Effects": "Motion Control\nMulti-elements to video\nExtend Video\nLip Sync\nAvatar\nText to Audio\nVideo to Audio\nText to Speech\nVoice Clone\nImage Recognize\nElement\nEffects\nEffect Templates\nNEW\nVideo Effects\nImage Generation\nModels\nImage Omni\nImage Generation\nReference to Image\nExtend Image\nAI Multi-Shot\nVirtual Try-On\nOthers\nQuery user info\nPricing\nBilling Info\nPrepaid Resource Packs\nProtocols\nPrivacy Policy of API Service\nTerms of API Service\nAPI Service Level Agreement\nVideo Effects\nCreate Task\nPOST\n/v1/videos/effects\ncURL\nCopy\nCollapse\ncurl --request POST \\\n --url https://api-singapore.klingai.com/v1/videos/effects \\\n --header 'Authorization: Bearer ' \\\n --header 'Content-Type: application/json' \\\n --data '{\n \"effect_scene\": \"color_mixing\",\n \"input\": {\n \"image\": \"https://p4-kling.klingai.com/kcdn/cdn-kcdn112452/kling-op/effects_raw_pic/color_mixing.jpeg\"\n }\n }'\n200\nCopy\nCollapse\n{\n \"code\": 0, //Error codes；Specific definitions can be found in \"Error Code\"\n \"message\": \"string\", //Error information\n \"request_id\": \"string\", //Request ID, generated by the system\n \"data\":{\n \"task_id\": \"string\", //Task ID, generated by the system\n \"task_status\": \"string\", //Task status, Enum values：submitted、processing、succeed、failed\n \"task_info\":{ //Task creation parameters\n \"external_task_id\": \"string\" //Customer-defined task ID\n },\n \"created_at\": 1722769557708, //Task creation time, Unix timestamp, unit ms\n \"updated_at\": 1722769557708 //Task update time, Unix timestamp, unit ms\n }\n}\n\nTotal of 229 video effects are available. You can achieve different effects by calling effect_scene. For detailed list, please refer to: Video Effects Center\n\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nRequest Body\neffect_scene\nstring\nRequired\n\nScene Name\n\nEnum values: flash_drive, shush_my_dreams, french_elegance, finger_swipe, advent_of_flora, smooth_transition, kiss_pro, raid_check, snow_night_kiss, eternal_kiss, fortune_in_motion, chinese_trend, sedan_chair_dance, skyfall, good_luck_dance, laicai_dance, yangge_dance, color_mixing, palm_sized_figure, lantern_festival_cuju, unique_firework, unique_spring_couplets, horse_mask, fortune_knocks_cartoon, tangyuan_to_animal, hot_feet_dance, swag_dance, pigeon_dance, bloodline_dance, chanel_dance, cute_dance, love_theme_song, pumpitup_dance, city_to_village, fortune_god_transform, new_year_feast, ring_in_new, horse_year_firework, pet_vlogger, crystal_horse, lateral_shift_transition, drunk_dance, drunk_dance_pet, daoma_dance, bouncy_dance, smooth_sailing_dance, new_year_greeting, lion_dance, prosperity, great_success, golden_horse_fortune, red_packet_box, lucky_horse_year, lucky_red_packet, lucky_money_come, lion_dance_pet, dumpling_making_pet, fish_making_pet, pet_red_packet, lantern_glow, expression_challenge, overdrive, heart_gesture_dance, poping, martial_arts, running, nezha, motorcycle_dance, subject_3_dance, ghost_step_dance, phantom_jewel, zoom_out, cheers_2026, fight_pro, hug_pro,heart_gesture_pro, dollar_rain_pro, pet_bee_pro, countdown_teleport, santa_random_surprise, magic_match_tree, bullet_time_360, happy_birthday, birthday_star, thumbs_up_pro, tiger_hug_pro, pet_lion_pro, surprise_bouquet, bouquet_drop, 3d_cartoon_1_pro, firework_2026, glamour_photo_shoot, box_of_joy, first_toast_of_the_year, my_santa_pic, santa_gift, steampunk_christmas, snowglobe, christmas_photo_shoot, ornament_crash​, santa_express, instant_christmas, particle_santa_surround, coronation_of_frost, building_sweater, spark_in_the_snow, scarlet_and_snow, cozy_toon_wrap, bullet_time_lite, magic_cloak, balloon_parade, jumping_ginger_joy, bullet_time, c4d_cartoon_pro, pure_white_wings, black_wings, golden_wing, pink_pink_wings, venomous_spider, throne_of_king, luminous_elf, woodland_elf, japanese_anime_1, american_comics, guardian_spirit, swish_swish, snowboarding, witch_transform, vampire_transform, pumpkin_head_transform, demon_transform, mummy_transform, zombie_transform, cute_pumpkin_transform, cute_ghost_transform, knock_knock_halloween, halloween_escape, baseball, inner_voice, a_list_look, memory_alive, trampoline, trampoline_night, pucker_up, guess_what, feed_mooncake, rampage_ape, flyer, dishwasher, pet_chinese_opera, magic_fireball, gallery_ring, pet_moto_rider, muscle_pet, squeeze_scream, pet_delivery, running_man, disappear, mythic_style, steampunk, 3d_cartoon_2, eagle_snatch, hug_from_past, firework, media_interview, pet_chef, santa_gifts, santa_hug, heart_gesture_1, pet_wizard, smoke_smoke, instant_kid, dollar_rain, cry_cry, building_collapse, gun_shot, mushroom, double_gun, pet_warrior, lightning_power, jesus_hug, shark_alert, long_hair, lie_flat, polar_bear_hug, brown_bear_hug , jazz_jazz, office_escape_plow, fly_fly, watermelon_bomb, pet_dance, boss_coming, wool_curly, pet_bee, marry_me, swing_swing, day_to_night, piggy_morph, wig_out, car_explosion, ski_ski, siblings, construction_worker, let’s_ride, snatched, magic_broom, felt_felt, jumpdrop, surfsurf, fairy_wing, angel_wing, dark_wing, skateskate, plushcut, jelly_press, jelly_slice, jelly_squish, jelly_jiggle, pixelpixel, yearbook, instant_film, anime_figure, rocketrocket, bloombloom, dizzydizzy, fuzzyfuzzy, squish, expansion, emoji\nFor more parameters, please refer to Video Effects Center\ninput\nobject\nRequired\n\nTask input structure. Fields vary depending on the scene.\n\nSingle-image effects (219 types) Scenes include: flash_drive, shush_my_dreams, advent_of_flora, raid_check, fortune_in_motion, chinese_trend, sedan_chair_dance, skyfall, good_luck_dance, laicai_dance, yangge_dance, color_mixing, palm_sized_figure, lantern_festival_cuju, unique_firework, unique_spring_couplets, horse_mask, fortune_knocks_cartoon, tangyuan_to_animal, hot_feet_dance, swag_dance, pigeon_dance, bloodline_dance, chanel_dance, cute_dance, love_theme_song, pumpitup_dance, city_to_village, fortune_god_transform, new_year_feast, ring_in_new, horse_year_firework, pet_vlogger, crystal_horse, lateral_shift_transition, drunk_dance, drunk_dance_pet, daoma_dance, bouncy_dance, smooth_sailing_dance, new_year_greeting, lion_dance, prosperity, great_success, golden_horse_fortune, red_packet_box, lucky_horse_year, lucky_red_packet, lucky_money_come, lion_dance_pet, dumpling_making_pet, fish_making_pet, pet_red_packet, lantern_glow, expression_challenge, overdrive, heart_gesture_dance, poping, martial_arts, running, nezha, motorcycle_dance, subject_3_dance, ghost_step_dance, phantom_jewel, zoom_out, dollar_rain_pro, pet_bee_pro, countdown_teleport, santa_random_surprise, magic_match_tree, bullet_time_360, happy_birthday, birthday_star, thumbs_up_pro, tiger_hug_pro, pet_lion_pro, surprise_bouquet, bouquet_drop, 3d_cartoon_1_pro, firework_2026, glamour_photo_shoot, box_of_joy, first_toast_of_the_year, my_santa_pic, santa_gift, steampunk_christmas, snowglobe, christmas_photo_shoot, ornament_crash​, santa_express, instant_christmas, particle_santa_surround, coronation_of_frost, building_sweater, spark_in_the_snow, scarlet_and_snow, cozy_toon_wrap, bullet_time_lite, magic_cloak, balloon_parade, jumping_ginger_joy, bullet_time, c4d_cartoon_pro, pure_white_wings, black_wings, golden_wing, pink_pink_wings, venomous_spider, throne_of_king, luminous_elf, woodland_elf, japanese_anime_1, american_comics, guardian_spirit, swish_swish, snowboarding, witch_transform, vampire_transform, pumpkin_head_transform, demon_transform, mummy_transform, zombie_transform, cute_pumpkin_transform, cute_ghost_transform, knock_knock_halloween, halloween_escape, baseball, inner_voice, a_list_look, memory_alive, trampoline, trampoline_night, pucker_up, guess_what, feed_mooncake, rampage_ape, flyer, dishwasher, pet_chinese_opera, magic_fireball, gallery_ring, pet_moto_rider, muscle_pet, squeeze_scream, pet_delivery, running_man, disappear, mythic_style, steampunk, 3d_cartoon_2, eagle_snatch, hug_from_past, firework, media_interview, pet_chef, santa_gifts, santa_hug, heart_gesture_1, pet_wizard, smoke_smoke, instant_kid, dollar_rain, cry_cry, building_collapse, gun_shot, mushroom, double_gun, pet_warrior, lightning_power, jesus_hug, shark_alert, long_hair, lie_flat, polar_bear_hug, brown_bear_hug , jazz_jazz, office_escape_plow, fly_fly, watermelon_bomb, pet_dance, boss_coming, wool_curly, pet_bee, marry_me, swing_swing, day_to_night, piggy_morph, wig_out, car_explosion, ski_ski, siblings, construction_worker, let’s_ride, snatched, magic_broom, felt_felt, jumpdrop, surfsurf, fairy_wing, angel_wing, dark_wing, skateskate, plushcut, jelly_press, jelly_slice, jelly_squish, jelly_jiggle, pixelpixel, yearbook, instant_film, anime_figure, rocketrocket, bloombloom, dizzydizzy, fuzzyfuzzy, squish, expansion, emoji\nDual-character effects (10 types available) Scenes include: french_elegance, finger_swipe, smooth_transition, kiss_pro, snow_night_kiss, eternal_kiss, cheers_2026, fight_pro, hug_pro, heart_gesture_pro\n▾\nHide child attributes\nimage\nstring\nOptional\n\nReference Image (for single-image effects)\n\nSupports image Base64 encoding or image URL (ensure accessibility)\nImportant: When using Base64, do NOT add any prefix like data:image/png;base64,. Submit only the raw Base64 string.\nCorrect Base64 format:\niVBORw0KGgoAAAANSUhEUgAAAAUA...\nIncorrect Base64 format (with data: prefix):\ndata:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAUA...\nSupported image formats: .jpg / .jpeg / .png\nFile size: ≤10MB, dimensions: ≥300px, aspect ratio: 1:2.5 ~ 2.5:1\nimages\narray\nOptional\n\nReference Image Group (for dual-character effects)\n\nArray length must be 2. The first image uploaded will be positioned on the left side of the composite photo, and the second image uploaded will be positioned on the right side.\nThis service includes composite photo functionality. Users upload two portrait images, and Kling AI will adaptively stitch them into a composite photo.\n\nSupports image Base64 encoding or image URL (ensure accessibility)\nImportant: When using Base64, do NOT add any prefix like data:image/png;base64,. Submit only the raw Base64 string.\nCorrect Base64 format:\niVBORw0KGgoAAAANSUhEUgAAAAUA...\nIncorrect Base64 format (with data: prefix):\ndata:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAUA...\nSupported image formats: .jpg / .jpeg / .png\nFile size: ≤10MB, dimensions: ≥300px, aspect ratio: 1:2.5 ~ 2.5:1\ncallback_url\nstring\nOptional\n\nThe callback notification address for the result of this task. If configured, the server will actively notify when the task status changes.\n\nFor specific message schema, see Callback Protocol\nexternal_task_id\nstring\nOptional\n\nCustomized Task ID\n\nUser-defined task ID. It will not override the system-generated task ID, but supports querying tasks by this ID\nPlease note that it must be unique for each user\nSingle-image Effect Request Example\n{\n \"effect_scene\": \"pet_lion\",\n \"input\":{\n \"image\":\"https://p4-kling.klingai.com/bs2/upload-ylab-stunt/c54e463c95816d959602f1f2541c62b2.png?x-kcdn-pid=112452\"\n }\n}\nDual-character Effect Request Example\n{\n \"effect_scene\": \"hug_pro\",\n \"input\": {\n \"images\": [\n \"https://example.com/image1.jpg\",\n \"https://example.com/image2.jpg\"\n ]\n }\n}\nQuery Task (Single)\nGET\n/v1/videos/effects/{id}\ncURL\nCopy\nCollapse\ncurl --request GET \\\n --url https://api-singapore.klingai.com/v1/videos/effects/{task_id} \\\n --header 'Authorization: Bearer ' \\\n --header 'Content-Type: application/json'\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes; Specific definitions can be found in Error codes\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system, is used to track requests and troubleshoot problems\n \"data\": {\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_status\": \"string\", // Task status, Enum values: submitted, processing, succeed, failed\n \"task_status_msg\": \"string\", // Task status information, d", "Query user info": "Motion Control\nMulti-elements to video\nExtend Video\nLip Sync\nAvatar\nText to Audio\nVideo to Audio\nText to Speech\nVoice Clone\nImage Recognize\nElement\nEffects\nEffect Templates\nNEW\nVideo Effects\nImage Generation\nModels\nImage Omni\nImage Generation\nReference to Image\nExtend Image\nAI Multi-Shot\nVirtual Try-On\nOthers\nQuery user info\nPricing\nBilling Info\nPrepaid Resource Packs\nProtocols\nPrivacy Policy of API Service\nTerms of API Service\nAPI Service Level Agreement\nAccount Information Inquiry\nQuery Resource Package List and Remaining Quantity under the Account\nGET\n/account/costs\ncURL\nPython\nCopy\nCollapse\ncurl --request GET \\\n --url 'https://api-singapore.klingai.com/account/costs?start_time=1726124664368&end_time=1727366400000' \\\n --header 'Authorization: Bearer ' \\\n --header 'Content-Type: application/json'\n200\nCopy\nCollapse\n{\n \"code\": 0, // Error codes; specific definitions see Error codes\n \"message\": \"string\", // Error information\n \"request_id\": \"string\", // Request ID, generated by the system, for tracking and troubleshooting\n \"data\": {\n \"code\": 0, // Error codes; specific definitions see Error codes\n \"msg\": \"string\", // Error information\n \"resource_pack_subscribe_infos\": [ // Resource package list\n {\n \"resource_pack_name\": \"Video Generation - 10,000 entries\", // Resource package name\n \"resource_pack_id\": \"509f3fd3d4ab4a3f9eec5db27aa44f27\", // Resource package ID\n \"resource_pack_type\": \"decreasing_total\", // Resource package type; \"decreasing_total\" = decreasing total, \"constant_period\" = constant periodicity\n \"total_quantity\": 200.0, // Total quantity\n \"remaining_quantity\": 118.0, // Remaining quantity (Please note: remaining quantity statistics have a 12h delay)\n \"purchase_time\": 1726124664368, // Purchase time, Unix timestamp in ms\n \"effective_time\": 1726124664368, // Effective time, Unix timestamp in ms\n \"invalid_time\": 1727366400000, // Expiration time, Unix timestamp in ms\n \"status\": \"expired\" // Resource package status: \"toBeOnline\" = Pending effectiveness, \"online\" = In effect, \"expired\" = Expired, \"runOut\" = Used up\n }\n ]\n }\n}\n\nNote: This API is free to call and allows you to query the list and balance of resource packages under your account. Please note to control the request rate (QPS<=1)\n\nRequest Header\nContent-Type\nstring\nRequired\nDefault to application/json\n\nData Exchange Format\n\nAuthorization\nstring\nRequired\n\nAuthentication information, refer to API authentication\n\nQuery Parameters\nstart_time\nint\nRequired\n\nQuery start time, Unix timestamp, unit: ms\n\nend_time\nint\nRequired\n\nQuery end time, Unix timestamp, unit: ms\n\nresource_pack_name\nstring\nOptional\n\nResource package name, used to precisely specify a resource package to query\n\nPrevious chapter：Virtual Try-On\nNext chapter：Billing Info\nThe Kling 3.0 Series Models API is Now Fully Available\n– All in One, One for All！\n\nModels Available in This Release\n\nKling 3.0 Motion Control, Kling Video 3.0, Kling Video 3.0 Omni, Kling Image 3.0, Kling Image 3.0 Omni\n\nRefer to \n\nKey Highlights of the Models\n\n3.0 All-in-One: A unified model for multi-modal input and output.\n\nMost powerful consistency across the universe: Subject consistency (supports cameo, subject with voice control, i2v + subject) and text consistency.\nNarrative control at your fingertips: More freedom, precision, and control—up to 15 seconds long, video scene cuts, ultra-high-definition storyboards/images, custom seconds.\nUpgraded native audio-visual output: Supports multiple speakers and languages (with accents).\n\nKling 3.0 Motion Control\n\nConsistent Facial Identity from any angle\nComplex Emotions faithfully reproduced\nHigh fidelity Restoration, Even with Face Occlusions\nConsistent Facial Clarity Across Dynamic Framing\n\nUser Guide ->\n\nKling Video 3.0\n\nCompared to 2.6, expected improvements:\n\nSupports subject upload in I2V scenarios for enhanced consistency\nSignificant improvement in multi-character referencing, especially for three-person scenarios\nSupports Japanese, Korean, and Spanish in addition to Chinese and English\nCapable of generating certain dialects and accents\nBetter distinction and control over different types of audio (speech, sound effects, BGM)\nImproved text retention in I2V scenarios\nSupports scene transitions, with up to 6 shots and customizable storyboarding\n\nUser Guide ->\n\nKling Video 3.0 Omni\n\nCompared to O1, expected improvements:\n\nNative audio-visual synchronization\nSupports video subject creation\nFurther improved consistency in reference-based tasks, especially for characters and products\nCombined capabilities of reference + storyboarding + audio-visual sync significantly enhance usability\nSupports scene transitions, with up to 6 shots\nExtended generation duration up to 15 seconds\n\nUser Guide ->\n\nKling Image 3.0\n\nHighly consistent feature retention\nPrecise response to detail modifications\nAccurate control over style and tone\nRich imaginative capabilities\n\nUser Guide ->\n\nKling Image 3.0 Omni\n\nEnhanced narrative sense\nNew storyboard image set generation, retaining reference image features with scene relevance\nDirect output of 2K/4K ultra-high-definition images\nFurther improved detail consistency\n\nUser Guide ->\n\nThank you for your support and understanding!\n\nI Got It", "Callback Schema": "Motion Control\nMulti-elements to video\nExtend Video\nLip Sync\nAvatar\nText to Audio\nVideo to Audio\nText to Speech\nVoice Clone\nImage Recognize\nElement\nEffects\nEffect Templates\nNEW\nVideo Effects\nImage Generation\nModels\nImage Omni\nImage Generation\nReference to Image\nExtend Image\nAI Multi-Shot\nVirtual Try-On\nOthers\nQuery user info\nPricing\nBilling Info\nPrepaid Resource Packs\nProtocols\nPrivacy Policy of API Service\nTerms of API Service\nAPI Service Level Agreement\nCallback Protocol\n\nAs for the Async task（image generation / video generation / virtual try-on），if you actively set the callback_url when you Create Task, the server will actively notify you when the task status changes, and the protocol is as follows:\n\nJSON\nCopy\nCollapse\n{\n \"task_id\": \"string\", // Task ID, generated by the system\n \"task_status\": \"string\", // Task status, Enum values: submitted, processing, succeed, failed\n \"task_status_msg\": \"string\", // Task status information, displaying the failure reason when the task fails (such as triggering the content risk control of the platform, etc.)\n \"created_at\": 1722769557708, // Task creation time, Unix timestamp, unit ms\n \"updated_at\": 1722769557708, // Task update time, Unix timestamp, unit ms\n \"final_unit_deduction\": \"string\", // The deduction units of task\n \"task_info\": { // Task creation parameters. Detailed information provided by the user during task creation.\n \"parent_video\": {\n \"id\": \"string\", // Generated video ID; globally unique\n \"url\": \"string\", // URL for generating images (To ensure information security, generated images/videos will be cleared after 30 days. Please make sure to save them promptly.)\n \"duration\": \"string\" // Total duration of the video before continuation, in s\n },\n \"external_task_id\": \"string\" // Customer-defined task ID\n },\n \"task_result\": {\n \"images\": [ // The result of image-related tasks\n {\n \"index\": int, // Image Number\n \"url\": \"string\" // URL for generating images, such as: https://h1.inkwai.com/bs2/upload-ylab-stunt/xxx.png (To ensure information security, generated images/videos will be cleared after 30 days. Please make sure to save them promptly.)\n }\n ],\n \"videos\": [ // The result of video-related tasks\n {\n \"id\": \"string\", // Generated video ID; globally unique\n \"url\": \"string\", // URL for generating videos (To ensure information security, generated images/videos will be cleared after 30 days. Please make sure to save them promptly.)\n \"duration\": \"string\" // Total video duration, unit: s (seconds)\n }\n ]\n }\n}\nPrevious chapter：Rate Limits\nNext chapter：Models\nThe Kling 3.0 Series Models API is Now Fully Available\n– All in One, One for All！\n\nModels Available in This Release\n\nKling 3.0 Motion Control, Kling Video 3.0, Kling Video 3.0 Omni, Kling Image 3.0, Kling Image 3.0 Omni\n\nRefer to \n\nKey Highlights of the Models\n\n3.0 All-in-One: A unified model for multi-modal input and output.\n\nMost powerful consistency across the universe: Subject consistency (supports cameo, subject with voice control, i2v + subject) and text consistency.\nNarrative control at your fingertips: More freedom, precision, and control—up to 15 seconds long, video scene cuts, ultra-high-definition storyboards/images, custom seconds.\nUpgraded native audio-visual output: Supports multiple speakers and languages (with accents).\n\nKling 3.0 Motion Control\n\nConsistent Facial Identity from any angle\nComplex Emotions faithfully reproduced\nHigh fidelity Restoration, Even with Face Occlusions\nConsistent Facial Clarity Across Dynamic Framing\n\nUser Guide ->\n\nKling Video 3.0\n\nCompared to 2.6, expected improvements:\n\nSupports subject upload in I2V scenarios for enhanced consistency\nSignificant improvement in multi-character referencing, especially for three-person scenarios\nSupports Japanese, Korean, and Spanish in addition to Chinese and English\nCapable of generating certain dialects and accents\nBetter distinction and control over different types of audio (speech, sound effects, BGM)\nImproved text retention in I2V scenarios\nSupports scene transitions, with up to 6 shots and customizable storyboarding\n\nUser Guide ->\n\nKling Video 3.0 Omni\n\nCompared to O1, expected improvements:\n\nNative audio-visual synchronization\nSupports video subject creation\nFurther improved consistency in reference-based tasks, especially for characters and products\nCombined capabilities of reference + storyboarding + audio-visual sync significantly enhance usability\nSupports scene transitions, with up to 6 shots\nExtended generation duration up to 15 seconds\n\nUser Guide ->\n\nKling Image 3.0\n\nHighly consistent feature retention\nPrecise response to detail modifications\nAccurate control over style and tone\nRich imaginative capabilities\n\nUser Guide ->\n\nKling Image 3.0 Omni\n\nEnhanced narrative sense\nNew storyboard image set generation, retaining reference image features with scene relevance\nDirect output of 2K/4K ultra-high-definition images\nFurther improved detail consistency\n\nUser Guide ->\n\nThank you for your support and understanding!\n\nI Got It", "Rate Limits": "Motion Control\nMulti-elements to video\nExtend Video\nLip Sync\nAvatar\nText to Audio\nVideo to Audio\nText to Speech\nVoice Clone\nImage Recognize\nElement\nEffects\nEffect Templates\nNEW\nVideo Effects\nImage Generation\nModels\nImage Omni\nImage Generation\nReference to Image\nExtend Image\nAI Multi-Shot\nVirtual Try-On\nOthers\nQuery user info\nPricing\nBilling Info\nPrepaid Resource Packs\nProtocols\nPrivacy Policy of API Service\nTerms of API Service\nAPI Service Level Agreement\nConcurrency Rules\nWhat is Kling API concurrency?\n\nKling API concurrency refers to the maximum number of generation tasks that an account can process in parallel at any given time. This capability is determined by the resource package. A higher concurrency level allows you to submit more API generation requests simultaneously (each call to the task creation interface initiates a new generation task).\n\n💡\n\nNotes\n\nThis only applies to the task creation interface; query interfaces do not consume concurrency.\nThis limitation concerns the number of concurrent tasks and is unrelated to Queries Per Second(QPS)— the system imposes no QPS limit.\nCore Rules\nDimension\tRule Description\nApplication Scope\tApplied at the account level. Calculated independently per resource pack type (video/image/virtual try-on). All API keys under the same account share the same concurrency quota.\nOccupancy Logic\tA task occupies concurrency from entering submitted status until completion (including failures). Released immediately after task ends.\nQuota Calculation\tDetermined by the highest concurrency value among all active resource packages of the same type. Example: If a 5-concurrency + 10-concurrency video package are both active → video concurrency capacity = 10\n\nSpecial Notes\n\nVideo / Virtual Try-on tasks: Each task occupies 1 concurrency.\nImage generation tasks: Concurrency used = the n value in the API request parameter. (Example: n = 9 → occupies 9 concurrency)\nOver-limit Error Mechanism\n\nWhen the number of running tasks reaches the concurrency limit, submitting a request will return an error.\n\nJSON\nCopy\nCollapse\n{\n\t\"code\": 1303,\n\t\"message\": \"parallel task over resource pack limit\",\n\t\"request_id\": \"9984d27b-a408-4073-ae28-17ca6a13622d\" //uuid\n}\nRecommended Approach\n\nSince this error is triggered by system load (not by parameter issues), it is recommended to:\n\nBackoff Retry Strategy: Use an exponential backoff algorithm to delay retries (recommended initial delay ≥ 1 second).\nQueue Management: Control the submission rate through a task queue and dynamically adapt to available concurrency.\nPrevious chapter：General Info\nNext chapter：Callback Schema\nWhat is Kling API concurrency?\nCore Rules\nOver-limit Error Mechanism\nRecommended Approach\nThe Kling 3.0 Series Models API is Now Fully Available\n– All in One, One for All！\n\nModels Available in This Release\n\nKling 3.0 Motion Control, Kling Video 3.0, Kling Video 3.0 Omni, Kling Image 3.0, Kling Image 3.0 Omni\n\nRefer to \n\nKey Highlights of the Models\n\n3.0 All-in-One: A unified model for multi-modal input and output.\n\nMost powerful consistency across the universe: Subject consistency (supports cameo, subject with voice control, i2v + subject) and text consistency.\nNarrative control at your fingertips: More freedom, precision, and control—up to 15 seconds long, video scene cuts, ultra-high-definition storyboards/images, custom seconds.\nUpgraded native audio-visual output: Supports multiple speakers and languages (with accents).\n\nKling 3.0 Motion Control\n\nConsistent Facial Identity from any angle\nComplex Emotions faithfully reproduced\nHigh fidelity Restoration, Even with Face Occlusions\nConsistent Facial Clarity Across Dynamic Framing\n\nUser Guide ->\n\nKling Video 3.0\n\nCompared to 2.6, expected improvements:\n\nSupports subject upload in I2V scenarios for enhanced consistency\nSignificant improvement in multi-character referencing, especially for three-person scenarios\nSupports Japanese, Korean, and Spanish in addition to Chinese and English\nCapable of generating certain dialects and accents\nBetter distinction and control over different types of audio (speech, sound effects, BGM)\nImproved text retention in I2V scenarios\nSupports scene transitions, with up to 6 shots and customizable storyboarding\n\nUser Guide ->\n\nKling Video 3.0 Omni\n\nCompared to O1, expected improvements:\n\nNative audio-visual synchronization\nSupports video subject creation\nFurther improved consistency in reference-based tasks, especially for characters and products\nCombined capabilities of reference + storyboarding + audio-visual sync significantly enhance usability\nSupports scene transitions, with up to 6 shots\nExtended generation duration up to 15 seconds\n\nUser Guide ->\n\nKling Image 3.0\n\nHighly consistent feature retention\nPrecise response to detail modifications\nAccurate control over style and tone\nRich imaginative capabilities\n\nUser Guide ->\n\nKling Image 3.0 Omni\n\nEnhanced narrative sense\nNew storyboard image set generation, retaining reference image features with scene relevance\nDirect output of 2K/4K ultra-high-definition images\nFurther improved detail consistency\n\nUser Guide ->\n\nThank you for your support and understanding!\n\nI Got It", "General Info": "Motion Control\nMulti-elements to video\nExtend Video\nLip Sync\nAvatar\nText to Audio\nVideo to Audio\nText to Speech\nVoice Clone\nImage Recognize\nElement\nEffects\nEffect Templates\nNEW\nVideo Effects\nImage Generation\nModels\nImage Omni\nImage Generation\nReference to Image\nExtend Image\nAI Multi-Shot\nVirtual Try-On\nOthers\nQuery user info\nPricing\nBilling Info\nPrepaid Resource Packs\nProtocols\nPrivacy Policy of API Service\nTerms of API Service\nAPI Service Level Agreement\nGeneral Information\nAPI Domain\nhttps://api-singapore.klingai.com\n💡\n\nNotice: The API endpoint for the new system has been updated from https://api.klingai.com to https://api-singapore.klingai.com. This API is suitable for users whose servers are located outside of China.\n\nAPI Authentication\nStep-1：Obtain AccessKey + SecretKey\nStep-2：Every time you request the API, you need to generate an API Token according to the fixed encryption method; put Authorization = Bearer in the Request Header\nEncryption Method：Follow JWT（Json Web Token, RFC 7519）standard\nJWT consists of three parts：Header、Payload、Signature\nPython\nJava\nCopy\nCollapse\nimport time\nimport jwt\n\nak = \"\" # fill access key\nsk = \"\" # fill secret key\n\ndef encode_jwt_token(ak, sk):\n headers = {\n \"alg\": \"HS256\",\n \"typ\": \"JWT\"\n }\n payload = {\n \"iss\": ak,\n \"exp\": int(time.time()) + 1800, # The valid time, in this example, represents the current time+1800s(30min)\n \"nbf\": int(time.time()) - 5 # The time when it starts to take effect, in this example, represents the current time -5s\n }\n token = jwt.encode(payload, sk, headers=headers)\n return token\n\nauthorization = encode_jwt_token(ak, sk)\nprint(authorization) # Printing the generated API_TOKEN\nStep-3: Use the API Token generated in Step 2 to assemble the Authorization and include it in the Request Header.\nAssembly format: Authorization = “Bearer XXX”, where XXX is the API Token generated in Step 2.\nNote: There should be a space between Bearer and XXX.\nError Code\nHTTP Status Code\tService Code\tDefinition of Service Code\tExplaination of Service Code\tSuggested Solutions\n200\t0\tRequest successful\t-\t-\n401\t1000\tAuthentication failed\tAuthentication failed\tCheck if the Authorization is correct\n401\t1001\tAuthentication failed\tAuthorization is empty\tFill in the correct Authorization in the Request Header\n401\t1002\tAuthentication failed\tAuthorization is invalid\tFill in the correct Authorization in the Request Header\n401\t1003\tAuthentication failed\tAuthorization is not yet valid\tCheck the start effective time of the token, wait for it to take effect or reissue\n401\t1004\tAuthentication failed\tAuthorization has expired\tCheck the validity period of the token and reissue it\n429\t1100\tAccount exception\tAccount exception\tVerifying account configuration information\n429\t1101\tAccount exception\tAccount in arrears (postpaid scenario)\tRecharge the account to ensure sufficient balance\n429\t1102\tAccount exception\tResource pack depleted or expired (prepaid scenario)\tPurchase additional resource packages, or activate the post-payment service (if available)\n403\t1103\tAccount exception\tUnauthorized access to requested resource, such as API/model\tVerifying account permissions\n400\t1200\tInvalid request parameters\tInvalid request parameters\tCheck whether the request parameters are correct\n400\t1201\tInvalid request parameters\tInvalid parameters, such as incorrect key or illegal value\tRefer to the specific information in the message field of the returned body and modify the request parameters\n404\t1202\tInvalid request parameters\tThe requested method is invalid\tReview the API documentation and use the correct request method\n404\t1203\tInvalid request parameters\tThe requested resource does not exist, such as the model\tRefer to the specific information in the message field of the returned body and modify the request parameters\n400\t1300\tTrigger strategy\tTrigger strategy of the platform\tCheck if any platform policies have been triggered\n400\t1301\tTrigger strategy\tTrigger the content security policy of the platform\tCheck the input content, modify it, and resend the request\n429\t1302\tTrigger strategy\tThe API request is too fast, exceeding the platform’s rate limit\tReduce the request frequency, try again later, or contact customer service to increase the limit\n429\t1303\tTrigger strategy\tConcurrency or QPS exceeds the prepaid resource package limit\tReduce the request frequency, try again later, or contact customer service to increase the limit\n429\t1304\tTrigger strategy\tTrigger the platform’s IP whitelisting policy\tContact customer service\n500\t5000\tInternal error\tServer internal error\tTry again later, or contact customer service\n503\t5001\tInternal error\tServer temporarily unavailable, usually due to maintenance\tTry again later, or contact customer service\n504\t5002\tInternal error\tServer internal timeout, usually due to a backlog\tTry again later, or contact customer service\nPrevious chapter：Changelog\nNext chapter：Rate Limits\nAPI Domain\nAPI Authentication\nError Code\nThe Kling 3.0 Series Models API is Now Fully Available\n– All in One, One for All！\n\nModels Available in This Release\n\nKling 3.0 Motion Control, Kling Video 3.0, Kling Video 3.0 Omni, Kling Image 3.0, Kling Image 3.0 Omni\n\nRefer to \n\nKey Highlights of the Models\n\n3.0 All-in-One: A unified model for multi-modal input and output.\n\nMost powerful consistency across the universe: Subject consistency (supports cameo, subject with voice control, i2v + subject) and text consistency.\nNarrative control at your fingertips: More freedom, precision, and control—up to 15 seconds long, video scene cuts, ultra-high-definition storyboards/images, custom seconds.\nUpgraded native audio-visual output: Supports multiple speakers and languages (with accents).\n\nKling 3.0 Motion Control\n\nConsistent Facial Identity from any angle\nComplex Emotions faithfully reproduced\nHigh fidelity Restoration, Even with Face Occlusions\nConsistent Facial Clarity Across Dynamic Framing\n\nUser Guide ->\n\nKling Video 3.0\n\nCompared to 2.6, expected improvements:\n\nSupports subject upload in I2V scenarios for enhanced consistency\nSignificant improvement in multi-character referencing, especially for three-person scenarios\nSupports Japanese, Korean, and Spanish in addition to Chinese and English\nCapable of generating certain dialects and accents\nBetter distinction and control over different types of audio (speech, sound effects, BGM)\nImproved text retention in I2V scenarios\nSupports scene transitions, with up to 6 shots and customizable storyboarding\n\nUser Guide ->\n\nKling Video 3.0 Omni\n\nCompared to O1, expected improvements:\n\nNative audio-visual synchronization\nSupports video subject creation\nFurther improved consistency in reference-based tasks, especially for characters and products\nCombined capabilities of reference + storyboarding + audio-visual sync significantly enhance usability\nSupports scene transitions, with up to 6 shots\nExtended generation duration up to 15 seconds\n\nUser Guide ->\n\nKling Image 3.0\n\nHighly consistent feature retention\nPrecise response to detail modifications\nAccurate control over style and tone\nRich imaginative capabilities\n\nUser Guide ->\n\nKling Image 3.0 Omni\n\nEnhanced narrative sense\nNew storyboard image set generation, retaining reference image features with scene relevance\nDirect output of 2K/4K ultra-high-definition images\nFurther improved detail consistency\n\nUser Guide ->\n\nThank you for your support and understanding!\n\nI Got It", "Billing Info": "Motion Control\nMulti-elements to video\nExtend Video\nLip Sync\nAvatar\nText to Audio\nVideo to Audio\nText to Speech\nVoice Clone\nImage Recognize\nElement\nEffects\nEffect Templates\nNEW\nVideo Effects\nImage Generation\nModels\nImage Omni\nImage Generation\nReference to Image\nExtend Image\nAI Multi-Shot\nVirtual Try-On\nOthers\nQuery user info\nPricing\nBilling Info\nPrepaid Resource Packs\nProtocols\nPrivacy Policy of API Service\nTerms of API Service\nAPI Service Level Agreement\nBilling\nBilling Method\nCurrently, only prepaid resource packages are available for purchase.\nThe resource packages are categorized into three types based on “capabilities”: video generation package, image generation package, and virtual try-on package.\nPrepaid Resource Packs\n\nNo credits will be deducted for failed tasks, regardless of the reason for failure, including failures due to content moderation policies\n\nPrevious chapter：Query user info\nNext chapter：Prepaid Resource Packs\nBilling Method\nPrepaid Resource Packs\nThe Kling 3.0 Series Models API is Now Fully Available\n– All in One, One for All！\n\nModels Available in This Release\n\nKling 3.0 Motion Control, Kling Video 3.0, Kling Video 3.0 Omni, Kling Image 3.0, Kling Image 3.0 Omni\n\nRefer to \n\nKey Highlights of the Models\n\n3.0 All-in-One: A unified model for multi-modal input and output.\n\nMost powerful consistency across the universe: Subject consistency (supports cameo, subject with voice control, i2v + subject) and text consistency.\nNarrative control at your fingertips: More freedom, precision, and control—up to 15 seconds long, video scene cuts, ultra-high-definition storyboards/images, custom seconds.\nUpgraded native audio-visual output: Supports multiple speakers and languages (with accents).\n\nKling 3.0 Motion Control\n\nConsistent Facial Identity from any angle\nComplex Emotions faithfully reproduced\nHigh fidelity Restoration, Even with Face Occlusions\nConsistent Facial Clarity Across Dynamic Framing\n\nUser Guide ->\n\nKling Video 3.0\n\nCompared to 2.6, expected improvements:\n\nSupports subject upload in I2V scenarios for enhanced consistency\nSignificant improvement in multi-character referencing, especially for three-person scenarios\nSupports Japanese, Korean, and Spanish in addition to Chinese and English\nCapable of generating certain dialects and accents\nBetter distinction and control over different types of audio (speech, sound effects, BGM)\nImproved text retention in I2V scenarios\nSupports scene transitions, with up to 6 shots and customizable storyboarding\n\nUser Guide ->\n\nKling Video 3.0 Omni\n\nCompared to O1, expected improvements:\n\nNative audio-visual synchronization\nSupports video subject creation\nFurther improved consistency in reference-based tasks, especially for characters and products\nCombined capabilities of reference + storyboarding + audio-visual sync significantly enhance usability\nSupports scene transitions, with up to 6 shots\nExtended generation duration up to 15 seconds\n\nUser Guide ->\n\nKling Image 3.0\n\nHighly consistent feature retention\nPrecise response to detail modifications\nAccurate control over style and tone\nRich imaginative capabilities\n\nUser Guide ->\n\nKling Image 3.0 Omni\n\nEnhanced narrative sense\nNew storyboard image set generation, retaining reference image features with scene relevance\nDirect output of 2K/4K ultra-high-definition images\nFurther improved detail consistency\n\nUser Guide ->\n\nThank you for your support and understanding!\n\nI Got It", "Prepaid Resource Packs": "Motion Control\nMulti-elements to video\nExtend Video\nLip Sync\nAvatar\nText to Audio\nVideo to Audio\nText to Speech\nVoice Clone\nImage Recognize\nElement\nEffects\nEffect Templates\nNEW\nVideo Effects\nImage Generation\nModels\nImage Omni\nImage Generation\nReference to Image\nExtend Image\nAI Multi-Shot\nVirtual Try-On\nOthers\nQuery user info\nPricing\nBilling Info\nPrepaid Resource Packs\nProtocols\nPrivacy Policy of API Service\nTerms of API Service\nAPI Service Level Agreement\nPrepaid Resource Packs\nImage Generation Resource Package\n\nCredit Deduction Details\n\nSingle Image Specification\n\n\t\n\nResource Package Deduction Count\n\n\t\n\nPricing (Original)\n\n【Kolors Image-O1】text-to-image\n\n\t\n\nDeduct 8 unit from total\n\n\t\n\n$0.028\n\n【Kolors Image-O1】image-to-image\n\n\t\n\nDeduct 8 unit from total\n\n\t\n\n$0.028\n\n【Kolors Image-O1】image editing\n\n\t\n\nDeduct 8 unit from total\n\n\t\n\n$0.028\n\n【Kolors Image-3O】1K/2K\n\n\t\n\nDeduct 8 unit from total\n\n\t\n\n$0.028\n\n【Kolors Image-3O】4K\n\n\t\n\nDeduct 16 unit from total\n\n\t\n\n$0.056\n\n【Kolors V1.0】text-to-image\n\n\t\n\nDeduct 1 unit from total\n\n\t\n\n$0.0035\n\n【Kolors V1.0】image-to-image\n\n\t\n\nDeduct 1 unit from total\n\n\t\n\n$0.0035\n\n【Kolors V1.5】text-to-image\n\n\t\n\nDeduct 4 units from total\n\n\t\n\n$0.014\n\n【Kolors V1.5】image-to-image\n\n\t\n\nDeduct 8 units from total\n\n\t\n\n$0.028\n\n【Kolors V2.0】text-to-image\n\n\t\n\nDeduct 4 units from total\n\n\t\n\n$0.014\n\n【Kolors V2.0】image-to-image\n\n\t\n\nDeduct 8 units from total\n\n\t\n\n$0.028\n\n【Kolors V2.0-new】image-to-image\n\n\t\n\nDeduct 8 units from total\n\n\t\n\n$0.028\n\n【Kolors V2.0】multi-image to image\n\n\t\n\nDeduct 16 units from total\n\n\t\n\n$0.056\n\n【Kolors V2.1】text-to-image\n\n\t\n\nDeduct 4 units from total\n\n\t\n\n$0.014\n\n【Kolors V2.1】multi-image to image\n\n\t\n\nDeduct 16 units from total\n\n\t\n\n$0.056\n\n【Kolors V3.0】1K/2K\n\n\t\n\nDeduct 8 unit from total\n\n\t\n\n$0.028\n\n【Image Editing】image expansion\n\n\t\n\nDeduct 8 units from total\n\n\t\n\n$0.028\n\n【AI Multi-Shot】Deduct based on the times of service access\n\n\t\n\nDeduct 20 units from total\n\n\t\n\n$0.07\n\nVideo Generation Resource Package\n\nCredit Deduction Details\n\nSingle Video Specification\n\n\t\n\nResource Package Deduction Count\n\n\t\n\nPricing (Original)\n\n【Kling Video-O1】\n\nStandard mode, 1-second video duration, without video input\n\n\t\n\nDeduct 0.6 units from total\n\n\t\n\n$0.084\n\n【Kling Video-O1】\n\nStandard mode, 1-second video duration, with video input\n\n\t\n\nDeduct 0.9 units from total\n\n\t\n\n$0.126\n\n【Kling Video-O1】\n\nProfessional mode, 1-second video duration, without video input\n\n\t\n\nDeduct 0.8 units from total\n\n\t\n\n$0.112\n\n【Kling Video-O1】\n\nProfessional mode, 1-second video duration, with video input\n\n\t\n\nDeduct 1.2 units from total\n\n\t\n\n$0.168\n\n【Kling Video-3O】Standard mode, 1-second video duration, without video input, without native audio generation\n\n\t\n\nDeduct 0.6 units from total\n\n\t\n\n$0.084\n\n【Kling Video-3O】Standard mode, 1-second video duration, without video input, with native audio generation\n\n\t\n\nDeduct 0.8 units from total\n\n\t\n\n$0.112\n\n【Kling Video-3O】Standard mode, 1-second video duration, with video input, without native audio generation\n\n\t\n\nDeduct 0.9 units from total\n\n\t\n\n$0.126\n\n【Kling Video-3O】Professional mode, 1-second video duration, without video input, without native audio generation\n\n\t\n\nDeduct 0.8 units from total\n\n\t\n\n$0.112\n\n【Kling Video-3O】Professional mode, 1-second video duration, without video input, with native audio generation\n\n\t\n\nDeduct 1.0 units from total\n\n\t\n\n$0.14\n\n【Kling Video-3O】Professional mode, 1-second video duration, with video input, without native audio generation\n\n\t\n\nDeduct 1.2 units from total\n\n\t\n\n$0.168\n\n【Kling V1.0】Standard mode, 5-second video duration\n\n\t\n\nDeduct 1 unit from total\n\n\t\n\n$0.14\n\n【Kling V1.0】Standard mode, 10-second video duration\n\n\t\n\nDeduct 2 units from total\n\n\t\n\n$0.28\n\n【Kling V1.0】Professional mode, 5-second video duration\n\n\t\n\nDeduct 3.5 units from total\n\n\t\n\n$0.49\n\n【Kling V1.0】Professional mode, 10-second video duration\n\n\t\n\nDeduct 7 units from total\n\n\t\n\n$0.98\n\n【Kling V1.5】Standard mode, 5-second video duration\n\n\t\n\nDeduct 2 units from total\n\n\t\n\n$0.28\n\n【Kling V1.5】Standard mode, 10-second video duration\n\n\t\n\nDeduct 4 units from total\n\n\t\n\n$0.56\n\n【Kling V1.5】Professional mode, 5-second video duration\n\n\t\n\nDeduct 3.5 units from total\n\n\t\n\n$0.49\n\n【Kling V1.5】Professional mode, 10-second video duration\n\n\t\n\nDeduct 7 units from total\n\n\t\n\n$0.98\n\n【Kling V1.6】Standard mode, 5-second video duration\n\n\t\n\nDeduct 2 units from total\n\n\t\n\n$0.28\n\n【Kling V1.6】Standard mode, 10-second video duration\n\n\t\n\nDeduct 4 units from total\n\n\t\n\n$0.56\n\n【Kling V1.6】Professional mode, 5-second video duration\n\n\t\n\nDeduct 3.5 units from total\n\n\t\n\n$0.49\n\n【Kling V1.6】Professional mode, 10-second video duration\n\n\t\n\nDeduct 7 units from total\n\n\t\n\n$0.98\n\n【Kling V1.6 multi-image to video】Standard mode, 5-second video duration\n\n\t\n\nDeduct 2 units from total\n\n\t\n\n$0.28\n\n【Kling V1.6 multi-image to video】Standard mode, 10-second video duration\n\n\t\n\nDeduct 4 units from total\n\n\t\n\n$0.56\n\n【Kling V1.6 multi-image to video】Professional mode, 5-second video duration\n\n\t\n\nDeduct 3.5 units from total\n\n\t\n\n$0.49\n\n【Kling V1.6 multi-image to video】Professional mode, 10-second video duration\n\n\t\n\nDeduct 7 units from total\n\n\t\n\n$0.98\n\n【Kling V2.0 Master】5-second video duration\n\n\t\n\nDeduct 10 units from total\n\n\t\n\n$1.4\n\n【Kling V2.0 Master】10-second video duration\n\n\t\n\nDeduct 20 units from total\n\n\t\n\n$2.8\n\n【Kling V2.1】Standard mode, 5-second video duration\n\n\t\n\nDeduct 2 units from total\n\n\t\n\n$0.28\n\n【Kling V2.1】Standard mode, 10-second video duration\n\n\t\n\nDeduct 4 units from total\n\n\t\n\n$0.56\n\n【Kling V2.1】Professional mode, 5-second video duration\n\n\t\n\nDeduct 3.5 units from total\n\n\t\n\n$0.49\n\n【Kling V2.1】Professional mode, 10-second video duration\n\n\t\n\nDeduct 7 units from total\n\n\t\n\n$0.98\n\n【Kling V2.1 Master】5-second video duration\n\n\t\n\nDeduct 10 units from total\n\n\t\n\n$1.4\n\n【Kling V2.1 Master】10-second video duration\n\n\t\n\nDeduct 20 units from total\n\n\t\n\n$2.8\n\n【Kling V2.5 Turbo】Standard mode, 5-second video duration\n\n\t\n\nDeduct 1.5 units from total\n\n\t\n\n$0.21\n\n【Kling V2.5 Turbo】Standard mode, 10-second video duration\n\n\t\n\nDeduct 3 units from total\n\n\t\n\n$0.42\n\n【Kling V2.5 Turbo】Professional mode, 5-second video duration\n\n\t\n\nDeduct 2.5 units from total\n\n\t\n\n$0.35\n\n【Kling V2.5 Turbo】Professional mode, 10-second video duration\n\n\t\n\nDeduct 5 units from total\n\n\t\n\n$0.7\n\n【Kling V2.6】Standard mode, 5-second video duration, without native audio generation\n\n\t\n\nDeduct 1.5 units from total\n\n\t\n\n$0.21\n\n【Kling V2.6】Standard mode, 10-second video duration, without native audio generation\n\n\t\n\nDeduct 3 units from total\n\n\t\n\n$0.42\n\n【Kling V2.6】Professional mode, 5-second video duration, without native audio generation\n\n\t\n\nDeduct 2.5 units from total\n\n\t\n\n$0.35\n\n【Kling V2.6】Professional mode, 10-second video duration, without native audio generation\n\n\t\n\nDeduct 5 units from total\n\n\t\n\n$0.7\n\n【Kling V2.6】Professional mode, 5-second video duration, with native audio generation, without voice control\n\n\t\n\nDeduct 5 units from total\n\n\t\n\n$0.7\n\n【Kling V2.6】Professional mode, 10-second video duration, with native audio generation, without voice control\n\n\t\n\nDeduct 10 units from total\n\n\t\n\n$1.4\n\n【Kling V2.6】Professional mode, 5-second video duration, with native audio generation, with voice control\n\n\t\n\nDeduct 6 units from total\n\n\t\n\n$0.84\n\n【Kling V2.6】Professional mode, 10-second video duration, with native audio generation, with voice control\n\n\t\n\nDeduct 12 units from total\n\n\t\n\n$1.68\n\n【Kling V3.0】Standard mode, 1-second video duration, without native audio generation\n\n\t\n\nDeduct 0.6 units from total\n\n\t\n\n$0.084\n\n【Kling V3.0】Standard mode, 1-second video duration, with native audio generation, without voice control\n\n\t\n\nDeduct 0.9 units from total\n\n\t\n\n$0.126\n\n【Kling V3.0】Professional mode, 1-second video duration, without native audio generation\n\n\t\n\nDeduct 0.8 units from total\n\n\t\n\n$0.112\n\n【Kling V3.0】Professional mode, 1-second video duration, with native audio generation, without voice control\n\n\t\n\nDeduct 1.2 units from total\n\n\t\n\n$0.168\n\n【Kling V2.6】Motion Control, Standard mode, per second video duration\n\n\t\n\nDeduct 0.5 unit from total\n\n\t\n\n$0.07\n\n【Kling V2.6】Motion Control, Pro mode, per second video duration\n\n\t\n\nDeduct 0.8 units from total\n\n\t\n\n$0.112\n\n【Kling V3.0】Motion Control, Standard mode, 1-second video duration\n\n\t\n\nDeduct 0.9 units from total\n\n\t\n\n$0.126\n\n【Kling V3.0】Motion Control, Professional mode, 1-second video duration\n\n\t\n\nDeduct 1.2 units from total\n\n\t\n\n$0.168\n\n【Kling V1.6】Multi-elements, Standard mode, 5-second video duration\n\n\t\n\nDeduct 3 units from total\n\n\t\n\n$0.42\n\n【Kling V1.6】Multi-elements, Standard mode, 10-second video duration\n\n\t\n\nDeduct 6 units from total\n\n\t\n\n$0.84\n\n【Kling V1.6】Multi-elements, Professional mode, 5-second video duration\n\n\t\n\nDeduct 5 units from total\n\n\t\n\n$0.7\n\n【Kling V1.6】Multi-elements, Professional mode, 10-second video duration\n\n\t\n\nDeduct 10 units from total\n\n\t\n\n$1.4\n\n【Kling V1.0】Video extension, Standard mode, 4~5-second video duration\n\n\t\n\nDeduct 1 unit from total\n\n\t\n\n$0.14\n\n【Kling V1.0】Video extension, Professional mode, 4~5-second video duration\n\n\t\n\nDeduct 3.5 units from total\n\n\t\n\n$0.49\n\n【Kling V1.5】Video extension, Standard mode, 4~5-second video duration\n\n\t\n\nDeduct 2 units from total\n\n\t\n\n$0.28\n\n【Kling V1.5】Video extension, Professional mode, 4~5-second video duration\n\n\t\n\nDeduct 3.5 units from total\n\n\t\n\n$0.49\n\n【Kling V1.6】Video extension, Standard mode, 4~5-second video duration\n\n\t\n\nDeduct 2 units from total\n\n\t\n\n$0.28\n\n【Kling V1.6】Video extension, Professional mode, 4~5-second video duration\n\n\t\n\nDeduct 3.5 units from total\n\n\t\n\n$0.49\n\n【Avatar】Standard mode, Charged by seconds, rounding up the visible duration, with the actual price rounded to the nearest second\n\n\t\n\nDeduct 0.4 unit from total\n\n\t\n\n$0.056\n\n【Avatar】Professional mode, Charged by seconds, rounding up the visible duration, with the actual price rounded to the nearest second\n\n\t\n\nDeduct 0.8 unit from total\n\n\t\n\n$0.112\n\n【Lip-sync】It is related to video duration, less than 5 seconds rounded up to 5 seconds\n\n\t\n\nDeduct 0.5 per 5 seconds from total\n\n\t\n\n$0.07\n\n【Video Effects】 relates to the template, and each video effect template has a different cost\n\n\t\n\nPrice List for Special Effects: Price List\n\n\t\n\nPrice List for Special Effects: Price List\n\n【Kling video-to-audio】 3~20s video duration\n\n\t\n\nDeduct 0.25 unit from total\n\n\t\n\n$0.035\n\n【Kling text-to-audio】 3~10s audio duration\n\n\t\n\nDeduct 0.25 unit from total\n\n\t\n\n$0.035\n\n【Identify Face】Deduct based on the times of service access\n\n\t\n\nDeduct 0.05 units from total\n\n\t\n\n$0.007\n\n【TTS】Deduct based on the times of service access\n\n\t\n\nDeduct 0.05 units from total\n\n\t\n\n$0.007\n\n【Image Recognize】Deduct based on the times of service access, include all sets of results\n\n\t\n\nDeduct 0.1 units from total\n\n\t\n\n$0.014\n\n【Custom Voice】Deduct based on the times of service access\n\n\t\n\nDeduct 0.05 units from total\n\n\t\n\n$0.007\n\nVirtual Try-on Resource Package\n\nCredit Deduction Details\n\nSingle Image Specification\n\n\t\n\nResource Package Deduction Count\n\n\t\n\nPricing (Original)\n\n【Kolors-Virtual-Try-on V1.0】\n\n\t\n\nDeduct 1 count from total\n\n\t\n\n$0.07\n\n【Kolors-Virtual-Try-on V1.5】\n\n\t\n\nDeduct 1 count from total\n\n\t\n\n$0.07\n\nPrevious chapter：Billing Info\nNext chapter：Privacy Policy of API Service\nThe Kling 3.0 Series Models API is Now Fully Available\n– All in One, One for All！\n\nModels Available in This Release\n\nKling 3.0 Motion Control, Kling Video 3.0, Kling Video 3.0 Omni, Kling Image 3.0, Kling Image 3.0 Omni\n\nRefer to \n\nKey Highlights of the Models\n\n3.0 All-in-One: A unified model for multi-modal input and output.\n\nMost powerful consistency across the universe: Subject consistency (supports cameo, subject with voice control, i2v + subject) and text consistency.\nNarrative control at your fingertips: More freedom, precision, and control—up to 15 seconds long, video scene cuts, ultra-high-definition storyboards/images, custom seconds.\nUpgraded native audio-visual output: Supports multiple speakers and languages (with accents).\n\nKling 3.0 Motion Control\n\nConsistent Facial Identity from any angle\nComplex Emotions faithfully reproduced\nHigh fidelity Rest" }