Qwen Image 2.1 VNCCS PoseStudio | 3D Pose Control
Load Diffusion Model
Already loads qwen_image_2.1_int8_convrot.safetensors
ComfyUI model file · Diffusion model (image)
Qwen Image 2.1 (int8 convrot) · Qwen Image 2.1 diffusion model for ComfyUI
This is Qwen Image 2.1’s diffusion model in int8 convrot, the half-size version of the bf16 weights that Comfy-Org repackaged for ComfyUI. It handles both text to image and image editing. Put it in ComfyUI/models/diffusion_models and load it with Load Diffusion Model.
You also need a Qwen3-VL 8B text encoder and qwen_image_2.1_vae_bf16.safetensors from the same repo. int8 convrot needs PyTorch with CUDA 13.
qwen_image_2.1_int8_convrot.safetensors
The short version
Qwen Image 2.1 is Qwen’s open image model for generation and editing. Comfy-Org repackages it as single files for ComfyUI: the diffusion model, a Qwen3-VL 8B text encoder, and a VAE.
This file is the diffusion model in int8 convrot, a rotation-based 8-bit format that halves the bf16 size (14.23 GB to 7.26 GB) with little quality loss. It is the base model, not Turbo, so it needs the usual step count unless you add a Turbo LoRA.
Install
ComfyUI repackage. Recommended.
saves as qwen_image_2.1_int8_convrot.safetensors
Same SHA256. May ask you to sign in.
saves as qwen_image_2.1_int8_convrot.safetensors
ComfyUI/ └─ models/ ├─ … └─ diffusion_models/ └─ qwen_image_2.1_int8_convrot.safetensors
$ cd ComfyUI/models/diffusion_models$ wget -O qwen_image_2.1_int8_convrot.safetensors \$ https://huggingface.co/Comfy-Org/Qwen-Image-2.1/resolve/main/diffusion_models/qwen_image_2.1_int8_convrot.safetensors$ sha256sum qwen_image_2.1_int8_convrot.safetensors# cb74113cb03faecd79611b01fd7fd642f0aa60d6f0b95086abee214d75eaa57d
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
Load Diffusion Model reads the diffusion model and outputs a MODEL for the sampler. The text encoder and VAE load in their own nodes.
Also load
A Qwen3-VL 8B text encoder (qwen3vl_8b_*) in Load CLIP and qwen_image_2.1_vae_bf16 in Load VAE.
Faster sampling
Use the Turbo model, or add a Turbo/5-8 step LoRA and set CFG to 1.
Works with
qwen_image_2.1_int8_convrot.safetensors is one part of the stack. A workflow that reports a missing model usually needs these too.
Run it
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.
7 workflows
These workflows already load qwen_image_2.1_int8_convrot.safetensors. Open one and run it as is.
Load Diffusion Model
Already loads qwen_image_2.1_int8_convrot.safetensors
Load Diffusion Model
Already loads qwen_image_2.1_int8_convrot.safetensors
Load Diffusion Model
Already loads qwen_image_2.1_int8_convrot.safetensors
Load Diffusion Model
Already loads qwen_image_2.1_int8_convrot.safetensors
Load Diffusion Model
Already loads qwen_image_2.1_int8_convrot.safetensors
Load Diffusion Model
Already loads qwen_image_2.1_int8_convrot.safetensors
Load Diffusion Model
Already loads qwen_image_2.1_int8_convrot.safetensors
1 workflow
These workflows load a different precision or format of the same model. Select qwen_image_2.1_int8_convrot.safetensors in their Load Diffusion Model node to use this version.
Load Diffusion Model
qwen_image_2.1_bf16.safetensors
Replace with qwen_image_2.1_int8_convrot.safetensors
2 workflows
These workflows load another Qwen Image 2.1 model. Select qwen_image_2.1_int8_convrot.safetensors in their Load Diffusion Model node to use it instead.
Load Diffusion Model
qwen_image_2.1_turbo_int8_convrot.safetensors
Replace with qwen_image_2.1_int8_convrot.safetensors
Load Diffusion Model
qwen_image_2.1_turbo_int8_convrot.safetensors
Replace with qwen_image_2.1_int8_convrot.safetensors
Compare
| File | Relation | Size | Best for |
|---|---|---|---|
| qwen_image_2.1_int8_convrot.safetensors | This page · base, int8 convrot | 7.26 GB | Default pick on PyTorch cu130 |
| qwen_image_2.1_bf16.safetensors | Same model · bf16 | 14.23 GB | Full precision, 24 GB+ GPUs |
| qwen_image_2.1_turbo_int8_convrot.safetensors | Turbo model · int8 convrot | 7.26 GB | About 8 steps, CFG 1 |
| qwenImage21INT4W4A8_int4W4A8.safetensors | Community · INT4 (W4A8) | 4.20 GB | Low VRAM |
| qwen21TurboOfficial_nvfp4.safetensors | Community · Turbo NVFP4 | 4.20 GB | RTX 50-series GPUs |
Troubleshoot
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.
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.
Why: int8 convrot needs a recent ComfyUI and a PyTorch build with CUDA 13 (cu130).
Fix: Update ComfyUI and PyTorch, or use qwen_image_2.1_bf16.safetensors.
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 7,256,783,064 bytes and that its SHA256 matches the value above. Download it again from Hugging Face · Comfy-Org/Qwen-Image-2.1 if not.
Rather skip the setup?
Run Qwen Image 2.1 VNCCS PoseStudio in your browser on a cloud GPU. Nothing to download.
FAQ
qwen3vl_8b_bf16, _int8_convrot or _w4a8) in models/text_encoders, and qwen_image_2.1_vae_bf16.safetensors in models/vae, all from Comfy-Org/Qwen-Image-2.1.ComfyUI/models/diffusion_models/. Then refresh the node list or restart ComfyUI so Load Diffusion Model lists the file.Sources: Comfy-Org/Qwen-Image-2.1 · Qwen/Qwen-Image-2.1 (original) · Civitai model 2954443
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.