QW

Qwen3.8 27B NVFP4 MTP SSMFIX

QwenGGUF by esatapedicoUpdated 2026-10-16
Download on Hugging Face
Parameters461M
Context256K
Layers65
Architectureqwen35
Smallest quant14.9 GB
Quick answer

Qwen3.8 27B NVFP4 MTP SSMFIX needs about 19.1 GB of memory at its main file (a 16.4 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 8 quants, from 14.9 GB to 23.2 GB. Longer context needs more: at 32K tokens the KV cache alone adds about 8.7 GB.

Qwen3.8-27B NVFP4 MTP SSMFIX – a dense vision‑language model with 27B parameters

Qwen3.8-27B NVFP4 MTP SSMFIX is a 27B‑parameter dense model that supports vision and long‑context reasoning. It is built on the Qwen3.5 architecture and includes an MTP speculative head and SSMFIX weight modification. The model can be run with vLLM or SGLang v0.5.19.

Good at

  • Vision‑language understanding
  • Long‑horizon agentic tasks
  • Coding and professional work

Limits

  • Requires more than 8 GB VRAM
  • No official benchmark scores
  • Experimental SSMFIX modification

Best for

  • Vision‑language inference
  • Agentic task planning
  • Code generation

Who it’s for. Developers needing a dense vision‑language model for complex, multi‑step tasks. The model is suitable for integration into existing stacks via vLLM or SGLang.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—14.9 GB 17.5 GB
Too big Get file →
—15.2 GB 17.8 GB
Too big Get file →
—15.5 GB 18.2 GB
Too big Get file →
—16.4 GB 19.1 GB
Too big Get file →
—16.9 GB 19.6 GB
Too big Get file →
—17.6 GB 20.3 GB
Too big Get file →
—19.7 GB 22.4 GB
Too big Get file →
—23.2 GB 25.9 GB
Too big Get file →

The eight NVFP4 GGUF files range from 14.86 GB to 23.19 GB. Estimated VRAM usage at 8 k context is 17.5 GB to 25.9 GB, so a GPU with at least 18 GB VRAM is recommended. For lower VRAM budgets, consider a smaller quant or a different model.

What GPU do you need for Qwen3.8 27B NVFP4 MTP SSMFIX?

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: no quant fits at an 8K context.
  • 24 GB VRAM: the best quant that fits is the main file (19.7 GB file, about 22.4 GB in use).
  • 32 GB VRAM: the best quant that fits is the main file (23.2 GB file, about 25.9 GB in use).

Which Qwen3.8 27B NVFP4 MTP SSMFIX quant should you choose?

If memory is tight, the smallest file is the first one in the table at 14.9 GB, with the quality trade-offs that come with fewer bits.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureqwen35
Parameters461M
Layers65
Trained context262,144 tokens
KV cache (fp16)260 KB per token

Source

Quantized byesatapedico
FamilyQwen
Licenseapache-2.0
Downloads (HF)2,802
Last updated2026-10-16

Frequently asked questions about Qwen3.8 27B NVFP4 MTP SSMFIX

How much VRAM does Qwen3.8 27B NVFP4 MTP SSMFIX need?

About 19.1 GB at its main file with an 8K context window. That is the 16.4 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 NVFP4 MTP SSMFIX on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 17.5 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Qwen3.8 27B NVFP4 MTP SSMFIX GGUF?

It depends on the quant: file is 14.9 GB, file is 15.2 GB, file is 15.5 GB, file is 16.4 GB, file is 16.9 GB, file is 17.6 GB and 2 more.

What context length does Qwen3.8 27B NVFP4 MTP SSMFIX support?

The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 2.2 GB at 8K and 8.7 GB at 32K.

Where can I download Qwen3.8 27B NVFP4 MTP SSMFIX GGUF?

The files are on Hugging Face in the esatapedico/Qwen3.8-27B-NVFP4-MTP-SSMFIX-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 NVFP4 MTP SSMFIX 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 8 GB VRAM?

No, the model requires at least 18 GB VRAM based on the estimated memory usage.

Does this model support vision?

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

What license does this model use?

The model is released under the apache-2.0 license.

Other GGUFs of Qwen3.8-27B-NVFP4

Similar models

Browse all →

More from esatapedico

All by esatapedico →

More Qwen models

All Qwen →

Other tiny models

All tiny models →