7 models · ranked by HuggingFace downloads
As an open-weight model, LTX-2.3 focuses on image to video. Training spans multiple languages, so LTX-2.3 covers cross-lingual image to video from one checkpoint. LTX-2.3 lists a non-standard license, so confirm permissions before deployment. Before relying on LTX-2.3, reproduce its key numbers on representative inputs.
LTX-2.3-fp8 is an open-weight checkpoint for image to video, distributed on the HuggingFace Hub. Prebuilt FP8 weights make local and edge inference of LTX-2.3-fp8 straightforward. LTX-2.3-fp8 is multilingual by design rather than English-only. LTX-2.3-fp8 is community-maintained, so track upstream changes and pin a known-good revision.
LTX-2 is an open-weight model aimed at image to video. LTX-2 lists a non-standard license, so confirm permissions before deployment. Training spans multiple languages, so LTX-2 covers cross-lingual image to video from one checkpoint. Before relying on LTX-2, reproduce its key numbers on representative inputs.
LTX-Video is Lightricks' open-source video generation model based on a DiT (Diffusion Transformer) architecture, generating video from text or image-plus-text prompts. It produces 768×512 video at 24fps and is notable for being among the first fully open-source high-quality video generation models with diffusers pipeline support. Lightricks targets creative professionals using video generation for commercial production.
As an open-weight model, Wan2.2-Distill-Loras focuses on image to video. The Apache 2.0 license keeps Wan2.2-Distill-Loras unrestricted for commercial reuse. Read Wan2.2-Distill-Loras's card for hardware requirements and licensing fine print before deploying.
LTX-2.3-GGUF targets image to video and is shipped as an open-weight, self-hostable checkpoint. Licensing for LTX-2.3-GGUF is unspecified or custom — clear it before commercial use. Prebuilt GGUF weights make local and edge inference of LTX-2.3-GGUF straightforward. Treat LTX-2.3-GGUF's published metrics as a starting point and validate against your workload.
stable-video-diffusion-img2vid-xt targets image to video and is shipped as an open-weight, self-hostable checkpoint. Licensing for stable-video-diffusion-img2vid-xt is unspecified or custom — clear it before commercial use. Evaluate stable-video-diffusion-img2vid-xt on your own data before trusting it in production.