ComfyUI model file · Diffusion model (image, community quant)

qwen21TurboOfficial_nvfp4.safetensors

Qwen Image 2.1 Turbo (NVFP4) · Qwen Image 2.1 Turbo in 4-bit NVFP4

qwen21TurboOfficial_nvfp4.safetensors is what Civitai names the NVFP4 file from the “Qwen 2.1 Turbo (Official)” listing. Inside is qwen_image_2.1_turbo_nvfp4.safetensors, Qwen Image 2.1 Turbo quantized to 4-bit NVFP4. Put it in ComfyUI/models/diffusion_models and load it with Load Diffusion Model.

  • Qwen Image 2.1 Turbo
  • NVFP4
  • 4.20 GB
  • Community upload

Despite “Official” in the listing name, this NVFP4 quant was uploaded by a community member (ferretduck), not by Qwen or Comfy-Org.

qwen21TurboOfficial_nvfp4.safetensors

Same file as
qwen_image_2.1_turbo_nvfp4.safetensors
Type
Diffusion model (image, community quant)
Format
Safetensors · NVFP4
Size
4.20 GB (4,197,594,329 bytes)
SHA256
b1cd960b…f0c720cfb
ComfyUI folder
models/diffusion_models/
Released
Oct 9, 2026 · Civitai upload
SHA256 and size from Civitai API (model 2993507, version 3394741), checked by RunComfy on .

The short version

What is qwen21TurboOfficial_nvfp4.safetensors?

Qwen Image 2.1 Turbo is the distilled version of Qwen Image 2.1 that generates in about 8 steps at CFG 1. Comfy-Org publishes it in bf16 and int8 convrot. This file is a community NVFP4 build of the same Turbo model, shrinking it to 4.20 GB.

NVFP4 is NVIDIA’s 4-bit floating-point format. It runs fastest on Blackwell GPUs (RTX 50-series and newer data-center cards), which have hardware support for it. On older GPUs, the int8 convrot Turbo from Comfy-Org is usually the safer choice.

The Civitai listing holds several Turbo builds (NVFP4, W4A8, W6A8, int8 convrot and bf16). Civitai renames each download after the listing and precision, which is how this file got the name qwen21TurboOfficial_nvfp4.

Install

Download and install qwen21TurboOfficial_nvfp4.safetensors

Where it goes
ComfyUI/
└─ models/
   ├─ …
   └─ diffusion_models/
      └─ qwen21TurboOfficial_nvfp4.safetensors
terminal · Linux / macOS
$ cd ComfyUI/models/diffusion_models# Download it from Civitai · Qwen 2.1 Turbo (Official), then save it here as:# qwen21TurboOfficial_nvfp4.safetensors$ sha256sum qwen21TurboOfficial_nvfp4.safetensors# b1cd960bb75161f04ed6cd399f3c248b826775793fab0c52bc217cdf0c720cfb

Safetensors files hold tensors only and cannot run code. Still compare the SHA256: a failed download often leaves an HTML error page saved under the model’s filename.

Use

How to load qwen21TurboOfficial_nvfp4.safetensors in ComfyUI

Load Diffusion Model reads the diffusion model and outputs a MODEL for the sampler. The text encoder and VAE load in their own nodes.

Minimal graph. Click a node to open its reference page.
  • Turbo settings

    About 8 steps with CFG 1. More steps or higher CFG tend to over-sharpen.

  • Also load

    A Qwen3-VL 8B text encoder (qwen3vl_8b_*) in Load CLIP and qwen_image_2.1_vae_bf16 in Load VAE.

Works with

Files a Qwen Image 2.1 workflow needs

qwen21TurboOfficial_nvfp4.safetensors is one part of the stack. A workflow that reports a missing model usually needs these too.

FileRoleFolderSizeDownload
qwen3vl_8b_int8_convrot.safetensorsText encoder (Qwen3-VL 8B)models/text_encoders/9.35 GB Get
qwen_image_2.1_vae_bf16.safetensorsVAEmodels/vae/676 MB Get

Run it

ComfyUI workflows that can use qwen21TurboOfficial_nvfp4.safetensors

We match workflows by the model files they load: this exact file first, then other versions of it, then workflows built for the same use, then other models from the same family.

  1. Uses another version of this file2
  2. Uses other Qwen Image 2.1 models8

Uses another version of this file

2 workflows

These workflows load a different precision or format of the same model. Select qwen21TurboOfficial_nvfp4.safetensors in their Load Diffusion Model node to use this version.

Uses other Qwen Image 2.1 models

8 workflows

These workflows load another Qwen Image 2.1 model. Select qwen21TurboOfficial_nvfp4.safetensors in their Load Diffusion Model node to use it instead.

Compare

Same file, other names, and alternatives

FileRelationSizeBest for
qwen21TurboOfficial_nvfp4.safetensorsThis page · Civitai download name4.20 GBRTX 50-series, low VRAM
qwen_image_2.1_turbo_nvfp4.safetensorsIdentical bytes · stored filename4.20 GBSame file
qwen_image_2.1_turbo_int8_convrot.safetensorsOfficial Turbo · int8 convrot7.26 GBMost GPUs on PyTorch cu130
qwen_image_2.1_int8_convrot.safetensorsBase model (not Turbo)7.26 GBFull step count, best quality

Troubleshoot

Common errors and fixes

Value not in list: unet_name: 'qwen21TurboOfficial_nvfp4.safetensors' not in [...]

Why: The workflow asks for this exact filename, but ComfyUI cannot find it in models/diffusion_models. Usually the file is missing, sits in a different folder, or was saved under another name such as qwen_image_2.1_turbo_nvfp4.safetensors.

Fix: Move the file into ComfyUI/models/diffusion_models/, rename it to match the workflow (or pick it again in the dropdown), then refresh the node list or restart ComfyUI.

SafetensorError: Error while deserializing header

Why: The download is incomplete, or the browser saved an HTML sign-in or error page under the model’s filename.

Fix: Check that the file is exactly 4,197,594,329 bytes and that its SHA256 matches the value above. Download it again from Civitai · Qwen 2.1 Turbo (Official) if not.

Rather skip the setup?

Run Qwen Image 2.1 Turbo Edit in your browser on a cloud GPU. Nothing to download.

Generate images with Qwen Image 2.1 online

FAQ

qwen21TurboOfficial_nvfp4.safetensors FAQ

Is this an official Qwen release?

No. The Civitai listing is called “Qwen 2.1 Turbo (Official)” because it repackages the official Turbo weights, but the NVFP4 quantization was uploaded by a community member. The official ComfyUI files are on Comfy-Org/Qwen-Image-2.1.

Does NVFP4 work on RTX 30 or 40 series cards?

NVFP4 is designed for Blackwell GPUs. On older cards it may not load or may run slower than int8. Use qwen_image_2.1_turbo_int8_convrot.safetensors instead.

Where do I put qwen21TurboOfficial_nvfp4.safetensors in ComfyUI?

In ComfyUI/models/diffusion_models/. Then refresh the node list or restart ComfyUI so Load Diffusion Model lists the file.

Which ComfyUI node loads qwen21TurboOfficial_nvfp4.safetensors?

Load Diffusion Model. It is built into ComfyUI, so no custom nodes are needed.

Can I use qwen21TurboOfficial_nvfp4.safetensors commercially?

Not by default. Qwen Image 2.1 is released under the Qwen Research License, which allows research and evaluation only. Commercial use needs a separate license from Qwen (model-business@notice.qwencloud.com). Community quantizations inherit the same terms.

Sources: Civitai model 2993507 · Comfy-Org/Qwen-Image-2.1 · Qwen/Qwen-Image-2.1 (original)

RunComfy
Copyright 2026 RunComfy. All Rights Reserved.

RunComfy is the premier ComfyUI platform, offering ComfyUI online environment and services, along with ComfyUI workflows featuring stunning visuals. RunComfy also provides AI Models, enabling artists to harness the latest AI tools to create incredible art.