kling-v2-6 is the final and most polished iteration of the second-generation Kling series by Kuaishou. It offers mature and highly stable image-to-video capabilities, featuring advanced motion control, superior text-video alignment, and cinematic aesthetics, bridging the gap to the next generation of video generation models. https://kling.ai/document-api/apiReference/model/imageToVideo
Generate video
All API requests must be authenticated using a Bearer token in the Authorization header. Please ensure your API key is active.Authorization: Bearer sk-xxxxxx
These parameters come from the selected model form_schema. Switching models updates this list and the request example.
Choose the generation mode. Text to Video: generate from storyboard prompts only. Image to Video: upload a first-frame image (optional last frame) as the start/end of the video, combined with storyboard prompts. Only one mode at a time.
image*string[Image to Video] The starting frame for video generation. JPG/PNG, ≤10MB, min 300px.
image_tail?string[Image to Video] Optional. End-frame control for precise first-to-last frame transition.
multi_prompt*arrayAdd multiple shots with unique prompts and durations. Total duration must match the setting below. Each prompt up to 2500 characters.
size?stringOutput video resolution. Default 1080P.
multi_shot?booleanWhen enabled, multiple storyboard shots can be controlled separately.
shot_type?stringCustomize: you set each shot's prompt and duration; Intelligence: the model plans the shot layout automatically.
watermark_enabled?booleanWhen enabled, the generated video carries a watermark. Disabled by default.
negative_prompt?stringDescribe content you don't want to see. Max 2500 characters.
mode?stringGeneration quality mode. Pro is higher quality but slower; Standard is faster. Pro is required when sound is enabled.
duration?numberTotal video duration in seconds. In multi-shot mode it must equal the sum of shot durations. For single-shot the standard values are 5 or 10 seconds.
cfg_scale?numberPrompt adherence, range 0-1, default 0.5. Higher values follow the prompt more strictly but reduce flexibility.
sound?stringWhether to generate native audio matching the visuals. Pro mode is required when on.
task_id?stringParameter description for Task Id
status?stringParameter description for Status
error?objectParameter description for Error
message?stringError Message
type?stringError Type
param?stringRelated Parameters
code?stringError Code