Qwen3.6 27b A2.8B needs about 13.0 GB of memory at its main file (a 12.0 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 2 quants, from 12.0 GB to 16.5 GB. Longer context needs more: at 32K tokens the KV cache alone adds about 2.1 GB.
Qwen3.6 27b A2.8B – a pruned, LoRA‑healed, GGUF model for fast decode on shared‑memory hardware
Qwen3.6 27b A2.8B is a pruned derivative of Qwen3.6‑35B‑A3B, built for fast decode on shared‑memory hardware. It removes 10 of 40 language‑model layers, applies LoRA healing, and is quantized to GGUF. The model supports multi‑token‑prediction for speculative decoding.
Good at
- Tool‑calling with MTP speculative decoding
- Fast decode on CPU/iGPU
- Large context support up to 262,144 tokens
Limits
- Requires llama.cpp build with MTP support
- Lowering routed‑expert count below 8 not recommended
- Speed‑demon build trades tool accuracy for size
Best for
- Agentic coding workflows
- Frontend workflow reasoning
- Tool‑enabled generation
Who it’s for. Developers needing a lightweight, fast‑decode version of Qwen3.6 for shared‑memory environments. The model is suitable for CPU or iGPU setups without discrete GPUs.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 12.0 GB | 13.0 GB | Too big | Get file → | |
| — | 16.5 GB | 17.5 GB | Too big | Get file → |
For a 12‑GB model file, expect about 13‑GB VRAM usage at 8‑K context. If you have 16‑GB VRAM, the 16.49‑GB file will use roughly 17.5‑GB. Choose the smaller file for tighter memory budgets; the larger file offers slightly more capacity for routed experts.
What GPU do you need for Qwen3.6 27b A2.8B?
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 (12.0 GB file, about 13.0 GB in use).
- 24 GB VRAM: the best quant that fits is the main file (16.5 GB file, about 17.5 GB in use).
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Qwen3.6 27b A2.8B
How much VRAM does Qwen3.6 27b A2.8B need?
About 13.0 GB at its main file with an 8K context window. That is the 12.0 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 Qwen3.6 27b A2.8B on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 13.0 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Qwen3.6 27b A2.8B GGUF?
It depends on the quant: file is 12.0 GB, file is 16.5 GB.
What context length does Qwen3.6 27b A2.8B support?
The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 0.5 GB at 8K and 2.1 GB at 32K.
Where can I download Qwen3.6 27b A2.8B GGUF?
The files are on Hugging Face in the Cyronius/Qwen3.6-27B-A2.8B repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does Qwen3.6 27b A2.8B 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 Qwen3.6 27b A2.8B on 8 GB VRAM?
The 12‑GB GGUF file estimates 13‑GB VRAM usage at 8‑K context, so 8 GB VRAM is insufficient. A larger VRAM or CPU/iGPU setup is required.
Does the model support vision?
No, the model is a causal language model with no vision encoder.
Is the model licensed for commercial use?
The model is licensed under apache‑2.0, which permits commercial use.