kwaivgi/kling-v1.6-standard 📝❓🔢🖼️ → 🖼️

⭐ Official ▶️ 1.3M runs 📅 Jan 2025 ⚙️ Cog 0.16.7
image-to-video text-to-video

About

Generate 5s and 10s videos in 720p resolution at 30fps

Example Output

Prompt:

"a portrait photo of a woman underwater with flowing hair"

Output

Performance Metrics

186.35s Prediction Time
219.83s Total Time
All Input Parameters
{
  "prompt": "a portrait photo of a woman underwater with flowing hair",
  "duration": 5,
  "cfg_scale": 0.5,
  "start_image": "https://replicate.delivery/pbxt/MNRKHnYUu5HjNqEerj2kxWRmUD3xWGaZ0gJmhqVbkra2jCbD/underwater.jpeg",
  "aspect_ratio": "16:9",
  "negative_prompt": ""
}
Input Parameters
prompt (required) Type: string
Text prompt for video generation
duration Default: 5
Duration of the video in seconds
cfg_scale Type: numberDefault: 0.5Range: 0 - 1
Flexibility in video generation; The higher the value, the lower the model's degree of flexibility, and the stronger the relevance to the user's prompt.
start_image Type: string
First frame of the video
aspect_ratio Default: 16:9
Aspect ratio of the video. Ignored if start_image is provided.
negative_prompt Type: stringDefault:
Things you do not want to see in the video
reference_images Type: array
Reference images to use in video generation (up to 4 images). Also known as scene elements.
Output Schema

Output

Type: stringFormat: uri

Example Execution Logs
Moderating content...

Moderating content...

Moderation complete in 0.25sec

Moderation complete in 1.82sec

Running prediction...
Using model: kling-v1-6 in std mode
Generating video...
Generated video in 181.6sec
Downloading 2847266 bytes
Downloaded 2.72MB in 2.53sec
Version Details
Version ID
a79b6f8e346976a7e3e85080c02076188a2bb9bf50b40fadcc996845ce826e73
Version Created
September 22, 2025
Run on Replicate →