YAML Metadata
Warning:
empty or missing yaml metadata in repo card
(https://huggingface.co/docs/hub/model-cards#model-card-metadata)
LTX2 Distilled Refine LoRA
This repo contains the distilled LoRA used for LTX-2 two-stage refinement.
FastVideo loads it only when ltx2_refine_enabled=True (or the equivalent refine
flags). The base model repo can reference this LoRA via
fastvideo_refine_lora_path in model_index.json, so users do not need to pass
the LoRA path manually.
Contents
ltx-2-19b-distilled-lora-384.safetensors— distilled LoRA weights for the stage-2 refine denoising pass.
Usage (FastVideo)
Use a distilled LTX-2 diffusers repo that includes
fastvideo_refine_lora_path="FastVideo/LTX2-Distilled-LoRA" in its
model_index.json. Then enable refinement:
from fastvideo import VideoGenerator
generator = VideoGenerator.from_pretrained(
"FastVideo/LTX2-Distilled-Diffusers",
ltx2_refine_enabled=True,
)
FastVideo will download this LoRA repo automatically only when refinement is enabled.
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
🙋
Ask for provider support