Lightricks/LTX-Video

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

10907 stars · 1121 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

Official repository for LTX-Video, a DiT-based open-weights video generation model from Lightricks that generates high-fidelity video (up to 4K, 50 FPS) from text, image, or audio inputs with synchronized audio. It provides inference code, ComfyUI and Diffusers integrations, and links to training tools, with development continuing in the LTX-2 repository.

Use cases

• generate videos from text prompts
• animate a still image into a video
• generate video with synchronized audio
• run video generation locally on my own GPU
• integrate video generation into ComfyUI workflows
• use a video generation model with Hugging Face diffusers
• fine-tune a video model on my own footage

When to choose

• you need open-weights video generation you can run or fine-tune yourself
• you want synchronized audio+video output in one model
• you need ComfyUI or Diffusers integration for video generation
• you want high-resolution (up to 4K) video generation with multiple performance modes

When to avoid

• you only need a hosted API without managing GPUs (use the LTX API instead)
• you need lightweight video editing rather than generative video creation
• you lack a capable GPU and cannot use cloud inference
• you need a closed-source commercial product with support guarantees

Facets

library · maturity active

video-processing machine-learning deep-learning llm-inference artificial-intelligence media deep-learning python cross-platform text-to-video image-to-video audio-to-video diffusion-transformer dit video-generation open-weights comfyui diffusers synchronized-audio video gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="Lightricks/LTX-Video")