Skip to content
Other

Wan-2.1-ComfyUI-Quants

Quantized Wan 2.1 Checkpoints for ComfyUI

1 min readupdated

This repository provides a collection of quantized (compressed) diffusion checkpoints and companion files for the Wan 2.1 model designed for ComfyUI. It includes models for text-to-video (T2V) and image-to-video (I2V 720p) generation, alongside a specialized SCAIL-2 checkpoint for motion transfer and optional LoRA adapters. The set is built for running local video generation with lower resource requirements.

Key features

  • Quantized versions of Wan 2.1 diffusion models (int8-lean-Convrot) for optimized performance
  • Includes checkpoints for text-to-video (T2V) and image-to-video (I2V 720p) generation
  • Additional SCAIL-2 checkpoint for motion transfer tasks
  • Set of helper files, including the UMT5-XXL text encoder and VAE

What you can do

  • Set up text-to-video generation in ComfyUI using quantized models
  • Run image-to-video (I2V 720p) generation using CLIP Vision H
  • Apply the SCAIL-2 checkpoint for pose and motion transfer via a custom workflow
  • Use optional LoRA adapters for DPO and relighting

Example

To get started, copy the model folders to the corresponding directories in your ComfyUI/models installation and select the required files in the interface loaders.

Read moreDetails
Contains quantized weights for Wan 2.1 (T2V and I2V 720p) sized around 13.5–15.5 GiB.
Requires UMT5-XXL (CLIPLoader type wan) and Wan 2.1 VAE.
Includes DPO and relight LoRAs at 1.143 GiB each, plus clip_vision_h.safetensors (1.177 GiB).
Discuss Telegram 0

Will it run?

Working out the memory needs

The bot is still collecting this model's weight sizes; the estimate will appear here in a few minutes. Meanwhile you can rent a GPU with headroom.

Massed ComputeOpen Massed Compute

Run it on Massed Compute

The script downloads the model to the rented machine. How to run it is in the model card.

  1. 1. Rent a machine with a fitting GPU

    Open Massed Compute
  2. 2. Paste the script into the machine's terminal

    Connect over SSH (the Massed Compute panel shows the command) or open a terminal on the machine's desktop.

    HF_TOKEN / CIVITAI_TOKENThe script starts with token lines: paste yours between the quotes. The Hugging Face token is for gated models, the Civitai one for Civitai models.

    # DmitryDB/Wan-2.1-ComfyUI-Quants
    # ===== Tokens: paste yours between the quotes =====
    export HF_TOKEN=""        # huggingface.co/settings/tokens, for gated models
    # ==========================================
    mkdir -p ~/models
    curl -LsSf https://astral.sh/uv/install.sh | sh
    source "$HOME/.local/bin/env"
    uv tool install "huggingface_hub[cli]"
    hf download DmitryDB/Wan-2.1-ComfyUI-Quants --local-dir ~/models/Wan-2.1-ComfyUI-Quants
    # The model is in ~/models. How to run it is in the model card (link above).
    Add more LoRAs to Wan-2.1-ComfyUI-Quants
  3. 3. Use it

    The model is in ~/models. How to run it is in the model card (link above).

Stop the machine in the panel when you are done: it bills by the minute while it runs.

How to speed it up

Picked for this model type. Choose your hardware above to see what fits you exactly.

  • FlashAttention 44.0.0b40from November 18, 2026new release

    fastest exact attention on B200

    Kernels for Blackwell (beta). Exact attention without quantisation — fine for everything, LLMs included.

    How to enable: pip install flash-attn-4; vLLM and SGLang pick it up themselves.

    Earlier generations (2)
    • FlashAttention 3 — The Hopper (H100/H200) version with FP8.
    • FlashAttention 2 — The standard for RTX 30/40/50 and A100: exact attention without the memory cost of long contexts.
  • torch.compile2.15.0from October 28, 2026new release

    10–30% for free

    Compiles the model for your GPU. Slower first run, faster afterwards.

    How to enable: model = torch.compile(model); in ComfyUI the TorchCompileModel node.

Model family

New posts about versions of Wan-2.1-ComfyUI-Quants (quants, finetunes, LoRAs) will show up under the bell.

No known relatives of this model yet. All derivatives on Hugging Face

About the model

DmitryDB/Wan-2.1-ComfyUI-Quants
Author
DmitryDB
Category
Other
Released
Posted

On NeyroLearn

Hugging Face

00

Ask AI about this model

Answered by Gemini 4 · the chat is kept only in this tab

Comments