QW

Qwen3.8 27B LowGPU NoMTP IQ3XXXS

QwenGGUF by TheWegemannUpdated 2026-08-22
Download on Hugging Face
Parameters26.9B
Context256K
Layers64
Architectureqwen35
Smallest quant9.6 GB
Quick answer

Qwen3.8 27B LowGPU NoMTP IQ3XXXS needs about 12.2 GB of memory at its main file (a 9.6 GB file) with an 8K context window, so it fits on a 16 GB GPU. Longer context needs more: at 32K tokens the KV cache alone adds about 8.6 GB.

Qwen3.8-27B LowGPU NoMTP IQ3XXXS – 27B dense model for 12 GB VRAM

Qwen3.8-27B LowGPU NoMTP IQ3XXXS is a 27 B dense model from the Qwen family. It is quantized with a custom per‑tensor mixed precision recipe to fit a 12 GB GPU while supporting full context and vision tasks. The model is released under the Apache‑2.0 license.

Good at

  • Vision‑language understanding
  • Long‑horizon reasoning
  • Agentic task completion

Limits

  • No MTP – single quant only
  • Requires ~12 GB VRAM at 8 k context
  • No official llama.cpp quant enum

Best for

  • Deploying on 12 GB GPUs
  • Running full‑GPU offload with large context
  • Using Hermes Agent with vision support

Who it’s for. Developers and researchers who need a compact Qwen3.8‑27B model that runs on a 12 GB GPU without MTP. The model is suitable for applications requiring vision, reasoning, and agentic capabilities within a single quantized file.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—9.6 GB 12.2 GB
Too big Get file →

The only available GGUF file is 9.57 GB with an estimated 12.2 GB memory usage at 8 k context. This is the recommended build for GPUs with 12 GB VRAM. No other quant options are provided, so use this file for inference.

What GPU do you need for Qwen3.8 27B LowGPU NoMTP IQ3XXXS?

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: no quant fits at an 8K context.
  • 12 GB VRAM: no quant fits at an 8K context.
  • 16 GB VRAM: the best quant that fits is the main file (9.6 GB file, about 12.2 GB in use).

Specifications

Architecture from the GGUF header

Architectureqwen35
Parameters26.9B
Layers64
Trained context262,144 tokens
KV cache (fp16)256 KB per token

Source

Quantized byTheWegemann
FamilyQwen
Licenseapache-2.0
Downloads (HF)642
Last updated2026-08-22

Frequently asked questions about Qwen3.8 27B LowGPU NoMTP IQ3XXXS

How much VRAM does Qwen3.8 27B LowGPU NoMTP IQ3XXXS need?

About 12.2 GB at its main file with an 8K context window. That is the 9.6 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 Qwen3.8 27B LowGPU NoMTP IQ3XXXS on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 12.2 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Qwen3.8 27B LowGPU NoMTP IQ3XXXS GGUF?

It depends on the quant: file is 9.6 GB.

What context length does Qwen3.8 27B LowGPU NoMTP IQ3XXXS support?

The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 2.1 GB at 8K and 8.6 GB at 32K.

Where can I download Qwen3.8 27B LowGPU NoMTP IQ3XXXS GGUF?

The files are on Hugging Face in the TheWegemann/Qwen3.8-27B-LowGPU-NoMTP-IQ3XXXS-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 Qwen3.8 27B LowGPU NoMTP IQ3XXXS 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 this model on a 12 GB GPU?

Yes, the 9.57 GB GGUF file is designed for 12 GB VRAM with full GPU offload and 8 k context.

Does this model support vision tasks?

Yes, it is a native vision‑language model that understands images and videos.

Is there a multi‑tensor product (MTP) version?

No, the model is released without MTP to save VRAM.

Other GGUFs of Qwen3.8-27B

Similar models

Browse all →

More from TheWegemann

All by TheWegemann →

More Qwen models

All Qwen →

Other medium models

All medium models →