kwaivgi/kling-v2.1 ❓📝🖼️ → 🖼️

⭐ Official ▶️ 2.2M runs 📅 Jun 2025 ⚙️ Cog 0.16.7
image-to-video

About

Use Kling v2.1 to generate 5s and 10s videos in 720p and 1080p resolution from a starting image (image-to-video)

Example Output

Prompt:

"a woman takes her hands out her pockets and gestures to the words with both hands, she is excited, behind her it is raining"

Output

Performance Metrics

125.23s Prediction Time
125.24s Total Time
All Input Parameters
{
  "mode": "standard",
  "prompt": "a woman takes her hands out her pockets and gestures to the words with both hands, she is excited, behind her it is raining",
  "duration": 5,
  "start_image": "https://replicate.delivery/xezq/rfKExHkg7L2UAyYNJj3p1YrW1M3ZROTQQXupJSOyM5RkwQcKA/tmpowaafuyw.png",
  "negative_prompt": ""
}
Input Parameters
mode Default: standard
Standard has a resolution of 720p, pro is 1080p. Both are 24fps.
prompt (required) Type: string
Text prompt for video generation
duration Default: 5
Duration of the video in seconds
end_image Type: string
Last frame of the video (pro mode is required when this parameter is set)
start_image (required) Type: string
First frame of the video. You must use a start image with kling-v2.1.
negative_prompt Type: stringDefault:
Things you do not want to see in the video
Output Schema

Output

Type: stringFormat: uri

Example Execution Logs
Using model: kling-v2-1 in std mode
Generating video...
Generated video in 122.1sec
Generated video ID: c8a352be-8fd6-44dc-b0ee-46773c12e63d
Downloading 3712142 bytes
Downloaded 3.54MB in 1.51sec
Version Details
Version ID
8f1d07f812d87339d7866c94ba2149e8ee456472e5c5ec04ac22795e21b55c68
Version Created
September 25, 2025
Run on Replicate →