LTX-Video text/image-to-video

Fast real-time-ish video generation with Lightricks LTX-Video 2B.

Estimated VRAM

~9.37 GB — ⚠️ ESTIMATE (unverified real-world usage). Estimate, not a measured figure: weight-size floor = ltx-video-2b-v0.9.safetensors (9.37 GB, sizes verified on Hugging Face/GitHub). Activations add more; ComfyUI offloading can reduce it. Source, checked 2026-10-13.

Models required

• ltx-video-2b-v0.9.safetensors — 9.37 GB ✅ verified 2026-10-13 (listing) · sha256 45369c2ae94e3949b127ef99ca9fe173b077f1ca0fecf8185e4115afcbc09581 → models/checkpoints
• t5xxl_fp16.safetensors — 9.788 GB ✅ verified 2026-10-13 (listing) · sha256 6e480b09fae049a72d2a8c5fbccb8d3e92febeb233bbe9dfe7256958a9167635 → models/text_encoders

Tested GPUs

No verified test reports yet.

Nodes

CheckpointLoaderSimple CLIPLoader LTXVConditioning EmptyLTXVLatentVideo SamplerCustom

Common errors & fixes

Wrong CLIP type

Set CLIPLoader type to 'ltxv'. ⚠️ unverified: Source page mentions the ltxv type, but the exact node setting is not confirmed.

Low-VRAM tips

⚠️ General guidance, not individually verified.

• fp8 T5
• Shorter clips

Source & credit

Workflow by comfyanonymous (ComfyUI_examples) / Lightricks. Get it from the original: https://comfyanonymous.github.io/ComfyUI_examples/ltxv/. License: LTX-Video open weights license (see Lightricks repo).