About LTX-2 Retake
LTX-2 Retake is a Multimodal LLM designed for targeted video editing. It excels in regenerating specific segments of existing videos (2-16 seconds) while preserving motion continuity, lighting, timing, and surrounding footage, enabling efficient revisions without full clip regeneration.
Some other noteworthy features of LTX-2 Retake include replace video only, replace audio only (e.g., dialogue or ambience), or replace both with natural language prompts; support for up to 20-second input clips at 1080p output; and seamless integration into iterative production workflows.
Ready to build with LTX-2 Retake?
Try LTX-2 Retake in the Workbench to prompt it, compare outputs, and iterate on prompts without writing any code. When you're ready to ship, call the same model from our API and build your own apps on top of it.
curl -sSf -X POST https://hub.oxen.ai/api/ai/videos/generate \
-H "Content-Type: application/json" \
-H "Authorization: Bearer $OXEN_API_KEY" \
-d '{
"model": "lightricks-ltx-2-retake",
"input_video": "https://hub.oxen.ai/api/repos/ox/Oxen-AI-Assets/file/main/videos/waterfall.mp4",
"prompt": "A cat sitting on a chair",
"start_time": 0,
"duration": 6,
"mode": "replace_video"
}'Request parameters
Every field LTX-2 Retake accepts in the request body. See the API reference for response formats.
https://hub.oxen.ai/api/ai/videos/generate| Field | Type | Required | Default | Description |
|---|---|---|---|---|
input_video | stringuri | Yes | — | Input video to edit, must be 1080p. |
prompt | string | Yes | — | Text prompt describing the edit |
start_time | number | No | 0 | Start time of the edit in seconds Range: 0 – +∞ |
duration | number | No | 6 | Range: 2 – 16 |
mode | string | No | "replace_video" | Mode of operationreplace_audioreplace_videoreplace_audio_and_video |