ドキュメントKling 2.5 Turbo
Model
kling-2.5-turbo
Single local model that automatically switches between text-to-video and image-to-video.
Model supplement for /unified-video/quote and /unified-video/generate権限: 実際の backend API key scope 設定に依存します
向いている用途
Teams that want one simple Kling option with duration-based pricing.
One model with two modes
Optional image input (1 max)
Simple duration pricing
What you can send
- Required: model, prompt
- Optional: images (0-1), aspectRatio, duration, negativePrompt, cfgScale, projectId
- Mode rule: 0 images = text-to-video, 1 image = image-to-video
Field rules
images
Optional. At most 1 image. Tail/last frame is not supported on this model.
aspectRatio
Supported: 1:1, 16:9, 9:16. Only used in text-to-video mode; ignored when an image is provided.
duration
Supported: 5, 10.
cfgScale
Clamped to 0..1.
negativePrompt
T2V max 2500 chars. I2V max 500 chars.
Pricing
Duration
Points
5s
35
10s
70
Frontend notes
- Always show prompt, duration, negativePrompt, and cfgScale.
- Show aspectRatio only in text-to-video mode.
- Treat image picker as a single optional input.
- Use callback as the main completion flow after generate.
すぐに試したいですか？
Settings で general API key を作成してから、ここに戻ってリクエスト例をコピーしてください。