LTXV2 Comfy
ltx-2-community-license-agreement
Model source
- Kind
- Other
- File size
- 22.4 MB–35.2 GB
- Updated
- Oct 20, 2026
- Publisher
- @Kijai
- Parameters
- 19B
- Source
- Hugging Face
Source excerpt
ltx-2-community-license-agreement
Sources
1 source
Sources
1 sourceModel artifacts
13 artifacts
Model artifacts
13 artifactsdiffusion_models/ltx-2-19b-dev_fp4_transformer_only.safetensors
safetensors · 13.5 GB · SHA-256 1428cd24d0e6…c756
diffusion_models/ltx-2-19b-dev_Q4_K_M.gguf
gguf · 11.8 GB · SHA-256 036e55c99980…9bf9
diffusion_models/ltx-2-19b-dev_Q8_0.gguf
gguf · 19.0 GB · SHA-256 0f1d67877689…c4a1
diffusion_models/ltx-2-19b-dev_transformer_only_bf16.safetensors
safetensors · 35.2 GB · SHA-256 3d387bfc694a…bd69
diffusion_models/ltx-2-19b-dev-fp8_transformer_only.safetensors
safetensors · 20.1 GB · SHA-256 6d6f26c0643e…6b76
diffusion_models/ltx-2-19b-distilled_transformer_only_bf16.safetensors
safetensors · 35.2 GB · SHA-256 2711d0565a6f…723f
diffusion_models/ltx-2-19b-distilled-fp8_transformer_only.safetensors
safetensors · 20.1 GB · SHA-256 ece636e1cb2a…43bb
loras/ltx-2-19b-distilled-lora_resized_dynamic_fro09_avg_rank_175_bf16.safetensors
safetensors · 3.33 GB · SHA-256 9bb035d2a267…ce05
loras/ltx-2-19b-distilled-lora_resized_dynamic_fro09_avg_rank_175_fp8.safetensors
safetensors · 1.67 GB · SHA-256 09b68d1df06c…b8c8
loras/ltx-2-19b-distilled-lora-resized_dynamic_fro095_avg_rank_242_bf16.safetensors
safetensors · 4.55 GB · SHA-256 252b5de7f7e1…7a6e
text_encoders/ltx-2-19b-embeddings_connector_dev_bf16.safetensors
safetensors · 2.67 GB · SHA-256 55781d7d90b6…fcc6
VAE/LTX2_video_vae_old_bf16.safetensors
safetensors · 2.32 GB · SHA-256 6bc274f9ca9c…f4a8
VAE/taeltx_2.safetensors
safetensors · 22.4 MB · SHA-256 6e4cc0469134…2f7b
Source excerpts
3 excerpts
Source excerpts
3 excerpts--- license: other license_name: ltx-2-community-license-agreement license_link: https://github.com/Lightricks/LTX-2/blob/main/LICENSE tags: - comfyui --- # 22th February 2026: Model loading change Starting from this commit https://github.com/Comfy-Org/ComfyUI/commit/f266b8d352607799afb4adf339cdfa854025185e Embedding connector has been moved from text encoder to diffusion model, so to continue using single files you now have to update KJNodes and use my loaders:  for GGUF:  # 22th January 2026: Tiny VAE and live previews: LTX2 Tiny VAE has been trained by [madebyollin](https://github.com/madebyollin), and now has preliminary support in ComfyUI nightly version. It can be used with normal VAE Loader and encode/decode nodes, but the quality is very low so it's only useful for preview purposes. Also currently for live animated sampler preview it can be only used with my **LTX2SamplingPreviewOverride** -node in KJNodes, simply load the VAE and plug it in, this overrides any preview setting too. <video controls autoplay width=50% src=https://cdn-uploads.huggingface.co/production/uploads/63297908f0b2fc94904a65b8/1nklT91omC_0yTSrC9x2D.mp4></video> # 13th of January 2026 update !!IMPORTANT!! Turns out the video VAE in the initial distilled checkpoints has been wrong one all this time, which (of course) was the one I initially extracted. It has now been replaced with the correct one, which should provide much higher detail ## requires using updated KJNodes or new enough core ComfyUI VAELoader to work correctly --- Separated LTX2 checkpoint for alternative way to load the models in Comfy  LTX2 GGUFs that include the metadata also work with the setup, if your GGUF nodes are up to date. Dev and distill embeddings_connector weights are extracted from their respective original models, they do give quite different results, but it's unclear if there's a "correct" one to use in any given situation. --- Loras folder includes rank reduced loras that can be useful to reduce the memory requirements.
Source context: 21434 downloads · 472 likes · Repo Kijai/LTXV2_comfy
Source context: 22385 downloads · 463 likes · Repo Kijai/LTXV2_comfy
