HunyuanVideo text-to-video

Tencent HunyuanVideo 13B text-to-video in ComfyUI.

Estimated VRAM

~45 GB @ 544x960x129f — ✅ verified official figure. Tencent README: minimum 45GB for 544x960x129f, 60GB for 720x1280x129f (reference implementation, not ComfyUI; FP8 weights save about 10GB). Source, checked 2026-10-13.

~60 GB @ 720x1280x129f — ✅ verified official figure. Tencent README: minimum 45GB for 544x960x129f, 60GB for 720x1280x129f (reference implementation, not ComfyUI; FP8 weights save about 10GB). Source, checked 2026-10-13.

Models required

• hunyuan_video_t2v_720p_bf16.safetensors — 25.642 GB ✅ verified 2026-10-13 (listing) · sha256 c6ff2d107f0fec571fe276ad847468404ed01855c28c0be8859c3b311daec52a → models/diffusion_models

Tested GPUs

No verified test reports yet.

Nodes

UNETLoader DualCLIPLoader EmptyHunyuanLatentVideo KSamplerAdvanced VAEDecodeTiled

Common errors & fixes

OOM during VAE decode

Use VAEDecodeTiled (used in the source workflow). ⚠️ unverified: Fix is general guidance; not confirmed against the cited source.

Low-VRAM tips

⚠️ General guidance, not individually verified.

• weight_dtype fp8
• Tiled VAE decode
• Fewer frames/lower res

Source & credit

Workflow by comfyanonymous (ComfyUI_examples) / Tencent. Get it from the original: https://comfyanonymous.github.io/ComfyUI_examples/hunyuan_video/. License: Tencent Hunyuan Community License.