QW

ComfyUI Qwen3 VL

QwenGGUF by makisekurisu-jpUpdated 2026-10-16
Download on Hugging Face
Parameters11.9B
Context256K
Layers36
Architectureqwen3vl
Smallest quant4.6 GB
Quick answer

ComfyUI Qwen3 VL needs about 6.8 GB of memory at IQ4_XS (a 5.1 GB file) with an 8K context window, so it fits on a 8 GB GPU. This repository has 3 quants, from 4.6 GB (IQ4_XS) to 6.6 GB (IQ4_XS). Longer context needs more: at 32K tokens the KV cache alone adds about 4.8 GB.

ComfyUI Qwen3 VL – 11.9B vision‑enabled model in GGUF format

ComfyUI Qwen3 VL is a 11.9B parameter vision‑enabled model based on the Qwen3VL architecture. It is distributed as GGUF files and can be run with llama.cpp. The model is licensed under apache-2.0.

Good at

  • Vision tasks
  • Multimodal inference
  • Image‑text understanding

Limits

  • No vision support in other quantizations
  • Requires 8k context for accurate memory estimates
  • No explicit language support listed

Best for

  • Image captioning
  • Visual question answering
  • Multimodal chat

Who it’s for. Developers and researchers needing a lightweight vision‑enabled LLM for local deployment. Suitable for projects that can compile llama‑cpp‑python and manage CUDA dependencies.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ4_XS4.6 GB 6.3 GB
Fits Get file →
IQ4_XS5.1 GB 6.8 GB
Fits Get file →
IQ4_XS6.6 GB 8.3 GB
Too big Get file →

For 8 GB VRAM, use the 4.56 GB IQ4_XS file (est. 6.3 GB memory). If 12 GB VRAM is available, the 5.07 GB IQ4_XS file (est. 6.8 GB memory) offers a balance of size and performance. For 16 GB VRAM, the 6.64 GB IQ4_XS file (est. 8.3 GB memory) provides the largest model variant.

What GPU do you need for ComfyUI Qwen3 VL?

These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: no quant fits at an 8K context.
  • 6 GB VRAM: no quant fits at an 8K context.
  • 8 GB VRAM: the best quant that fits is IQ4_XS (5.1 GB file, about 6.8 GB in use).
  • 12 GB VRAM: the best quant that fits is IQ4_XS (6.6 GB file, about 8.3 GB in use).

Which ComfyUI Qwen3 VL quant should you choose?

IQ4_XS (5.1 GB) is the middle option in this repository. The usual sweet spot between file size and quality, and the quant most people start with.

If memory is tight, the smallest file is IQ4_XS at 4.6 GB, with the quality trade-offs that come with fewer bits. The usual sweet spot between file size and quality, and the quant most people start with.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureqwen3vl
Parameters11.9B
Layers36
Trained context262,144 tokens
KV cache (fp16)144 KB per token

Source

Quantized bymakisekurisu-jp
FamilyQwen
Licenseapache-2.0
Downloads (HF)1,372
Last updated2026-10-16

Frequently asked questions about ComfyUI Qwen3 VL

How much VRAM does ComfyUI Qwen3 VL need?

About 6.8 GB at IQ4_XS with an 8K context window. That is the 5.1 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run ComfyUI Qwen3 VL on an 8 GB GPU?

Yes, with IQ4_XS: the 5.1 GB file needs about 6.8 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of ComfyUI Qwen3 VL GGUF?

It depends on the quant: IQ4_XS is 4.6 GB, IQ4_XS is 5.1 GB, IQ4_XS is 6.6 GB.

What context length does ComfyUI Qwen3 VL support?

The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 1.2 GB at 8K and 4.8 GB at 32K.

Where can I download ComfyUI Qwen3 VL GGUF?

The files are on Hugging Face in the makisekurisu-jp/ComfyUI-Qwen3-VL-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does ComfyUI Qwen3 VL use?

Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Can I run ComfyUI Qwen3 VL on 8 GB VRAM?

Yes, the 4.56 GB IQ4_XS file uses approximately 6.3 GB memory at 8k context, fitting within 8 GB VRAM.

Do I need to compile llama‑cpp‑python manually?

No, you can use JamePeng’s precompiled WHL package, ensuring the CUDA version matches exactly.

What license applies to this model?

The model is released under the apache-2.0 license.

Similar models

Browse all →

More from makisekurisu-jp

All by makisekurisu-jp →

More Qwen models

All Qwen →

Other medium models

All medium models →