Douravita UGC — LTX 2.3 I2V model bundle

Consolidated set of the 10 models needed to run the LTX 2.3 All-In-One workflow (mode 03 Image-to-Video with native audio). Re-packaged from upstream Hugging Face repos for a single-shot ~42 GB download during Docker builds.

Layout

diffusion_models/ltx-2.3-22b-dev_transformer_only_fp8_scaled.safetensors  (Kijai/LTX2.3_comfy)
vae/LTX23_video_vae_bf16.safetensors                                       (Kijai/LTX2.3_comfy)
vae/LTX23_audio_vae_bf16.safetensors                                       (Kijai/LTX2.3_comfy)
vae/taeltx2_3.safetensors                                                  (Kijai/LTX2.3_comfy)
text_encoders/gemma_3_12B_it_fp4_mixed.safetensors                         (Comfy-Org/ltx-2)
text_encoders/ltx-2.3_text_projection_bf16.safetensors                     (Kijai/LTX2.3_comfy)
latent_upscale_models/ltx-2.3-spatial-upscaler-x2-1.1.safetensors          (Lightricks/LTX-2.3)
loras/ltx-2.3-22b-distilled-lora-dynamic_fro09_avg_rank_105_bf16.safetensors (Kijai/LTX2.3_comfy)
loras/ltx-2.3-22b-ic-lora-union-control-ref0.5.safetensors                 (Lightricks/...)
loras/ltx-2-19b-ic-lora-detailer.safetensors                               (Lightricks/...)

Usage

hf download DouraVITA/ltx-ugc-bundle --local-dir ./models

All file paths are preserved — drop into ComfyUI/models/ directly.

Licenses

Each model carries its original upstream license. See source repos linked above.

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support