Qwen3.8-27B needs about 18.0 GB of memory at Q4_K (a 15.4 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 7 quants, from 11.2 GB to 16.4 GB (Q4_K). Longer context needs more: at 32K tokens the KV cache alone adds about 8.7 GB.
Qwen3.8-27B GGUF with non‑uniform GSQ‑RCO quantization and vision projector
This GGUF package contains Qwen3.8‑27B quantized with GSQ‑RCO, assigning per‑tensor precision to meet a size budget. The model includes a BF16 vision projector for multimodal use and optional Multi‑Token Prediction head. It runs unchanged in llama.cpp, Ollama, and LM Studio.
Good at
- Non‑uniform per‑tensor quantization
- Vision projector support
- Multi‑Token Prediction head
Limits
- Requires Intel Arc GPU for best performance
- No explicit vision token limits
- License is Apache‑2.0
Best for
- Local inference on 27B models
- Multimodal applications with vision encoder
- Speculative decoding with MTP head
Who it’s for. Developers and researchers needing a 27B LLM with efficient per‑tensor quantization and vision capabilities. Suitable for local deployment on systems with sufficient VRAM and Intel Arc GPUs.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 11.2 GB | 13.9 GB | Too big | Get file → | |
| — | 12.8 GB | 15.4 GB | Too big | Get file → | |
| Q4_K | 14.7 GB | 17.4 GB | Too big | Get file → | |
| Q4_K | 15.4 GB | 18.0 GB | Too big | Get file → | |
| IQ3_S | 15.4 GB | 18.0 GB | Too big | Get file → | |
| Q4_K | 16.2 GB | 18.9 GB | Too big | Get file → | |
| Q4_K | 16.4 GB | 19.1 GB | Too big | Get file → |
For 8 GB VRAM, use the Q4_K quantization (~17–19 GB memory at 8k context). For 16 GB VRAM, the IQ3_S variant (~18 GB) or larger Q4_K (~18.9–19.1 GB) are suitable. The unknown quant files are larger and not recommended for limited VRAM.
What GPU do you need for Qwen3.8-27B?
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 (12.8 GB file, about 15.4 GB in use).
- 24 GB VRAM: the best quant that fits is Q4_K (16.4 GB file, about 19.1 GB in use).
Which Qwen3.8-27B quant should you choose?
Q4_K (15.4 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 the first one in the table at 11.2 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
Source
Frequently asked questions about Qwen3.8-27B
How much VRAM does Qwen3.8-27B need?
About 18.0 GB at Q4_K with an 8K context window. That is the 15.4 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 on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 13.9 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Qwen3.8-27B GGUF?
It depends on the quant: file is 11.2 GB, file is 12.8 GB, Q4_K is 14.7 GB, Q4_K is 15.4 GB, IQ3_S is 15.4 GB, Q4_K is 16.2 GB and 1 more.
What context length does Qwen3.8-27B 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 GGUF?
The files are on Hugging Face in the Greatjedi/Qwen3.8-27B-Intel-Arc-Tuned-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 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 the Qwen3.8-27B GGUF on 8 GB VRAM?
The Q4_K variants require about 17–19 GB of VRAM at 8k context, so 8 GB is insufficient.
Does this model support vision tasks?
Yes, it includes a BF16 vision projector for multimodal use.
What license applies to this model?
The model is released under the Apache‑2.0 license.