Qwen3.6 28B REAP20 A3B Merged Bf16 needs about 18.4 GB of memory at Q4_K_M (a 17.3 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 12 quants, from 10.6 GB (Q2_K) to 56.5 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 2.7 GB.
Qwen3.6‑28B REAP20 A3B – 20% expert‑pruned, LoRA‑fine‑tuned Qwen3.6‑35B‑A3B
Qwen3.6‑28B REAP20 A3B is a 28.2 billion‑parameter model pruned by 20 % using REAP. It retains ~3 billion active parameters per token and is fine‑tuned on Claude Opus reasoning traces. The model is distributed in GGUF format with multiple quantizations.
Good at
- Reasoning with chain‑of‑thought
- Coding tasks
- Knowledge retrieval
Limits
- No vision support
- Limited to 262,144‑token context
- Requires Apache‑2.0 attribution
Best for
- Local inference on mid‑range GPUs
- Fine‑tuning on custom reasoning data
- Deploying in low‑latency API servers
Who it’s for. Developers needing a compact Qwen variant for reasoning and coding tasks. The model is suitable for users who can run GGUF files locally and want a balance between size and performance.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q2_K | 10.6 GB | 11.8 GB | Too big | Get file → | |
| IQ3_XXS | 11.2 GB | 12.3 GB | Too big | Get file → | |
| Q3_K_S | 12.4 GB | 13.6 GB | Too big | Get file → | |
| Q3_K_M | 13.7 GB | 14.9 GB | Too big | Get file → | |
| Q3_K_L | 14.8 GB | 16.0 GB | Too big | Get file → | |
| Q4_K_S | 16.2 GB | 17.4 GB | Too big | Get file → | |
| Q4_K_M | 17.3 GB | 18.4 GB | Too big | Get file → | |
| Q5_K_S | 19.6 GB | 20.7 GB | Too big | Get file → | |
| Q5_K_M | 20.2 GB | 21.3 GB | Too big | Get file → | |
| Q6_K | 23.2 GB | 24.4 GB | Too big | Get file → | |
| Q8_0 | 30.1 GB | 31.2 GB | Too big | Get file → | |
| BF16 | 56.5 GB | 57.7 GB | Too big | Get file → |
For 8 GB VRAM, use Q4_K_M (file size 17.26 GB, estimated 18.4 GB memory). For 16 GB VRAM, Q5_K_M (20.16 GB, 21.3 GB) or Q4_K_S (16.23 GB, 17.4 GB). For 24 GB VRAM, Q6_K (23.23 GB, 24.4 GB).
What GPU do you need for Qwen3.6 28B REAP20 A3B Merged Bf16?
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: the best quant that fits is Q2_K (10.6 GB file, about 11.8 GB in use).
- 16 GB VRAM: the best quant that fits is Q3_K_L (14.8 GB file, about 16.0 GB in use).
- 24 GB VRAM: the best quant that fits is Q5_K_M (20.2 GB file, about 21.3 GB in use).
- 32 GB VRAM: the best quant that fits is Q8_0 (30.1 GB file, about 31.2 GB in use).
- 48 GB VRAM: the best quant that fits is Q8_0 (30.1 GB file, about 31.2 GB in use).
Which Qwen3.6 28B REAP20 A3B Merged Bf16 quant should you choose?
Q4_K_M is the usual starting point for Qwen3.6 28B REAP20 A3B Merged Bf16: it balances file size and quality well. Here it is a 17.3 GB file.
The Q8_0 file is 30.1 GB, about 1.7 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.
If memory is tight, the smallest file is Q2_K at 10.6 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Qwen3.6 28B REAP20 A3B Merged Bf16
How much VRAM does Qwen3.6 28B REAP20 A3B Merged Bf16 need?
About 18.4 GB at Q4_K_M with an 8K context window. That is the 17.3 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 28B REAP20 A3B Merged Bf16 on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 11.8 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Qwen3.6 28B REAP20 A3B Merged Bf16 GGUF?
It depends on the quant: Q2_K is 10.6 GB, IQ3_XXS is 11.2 GB, Q3_K_S is 12.4 GB, Q3_K_M is 13.7 GB, Q3_K_L is 14.8 GB, Q4_K_S is 16.2 GB and 6 more.
What context length does Qwen3.6 28B REAP20 A3B Merged Bf16 support?
The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 0.7 GB at 8K and 2.7 GB at 32K.
Where can I download Qwen3.6 28B REAP20 A3B Merged Bf16 GGUF?
The files are on Hugging Face in the barozp/Qwen3.6-28B-REAP20-A3B-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 Qwen3.6 28B REAP20 A3B Merged Bf16 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 Qwen3.6‑28B REAP20 A3B on 8 GB VRAM?
Yes, using the Q4_K_M quantization which requires about 18.4 GB of VRAM at 8 k context.
Does this model support image input?
No, the model has no vision capabilities.
What license applies to this model?
Apache‑2.0, inherited from the base model; attribution is required.