Qwen3.6 29B REAP Opus Reasoning Distill MTP needs about 18.9 GB of memory at Q4_K_M (a 17.7 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 14 quants, from 7.2 GB (IQ1_M) to 57.9 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 2.8 GB.
Qwen3.6 29B REAP Opus Reasoning Distill MTP – 28.9B parameters, 262,144 token context
A 28.9B parameter Qwen model with REAP pruning, reasoning LoRA distillation, and a Multi‑Token Prediction head. It supports 262,144 token context and is optimized for self‑speculative decoding in llama.cpp.
Good at
- Reasoning tasks
- Tool use
- Self‑speculative decoding
Limits
- Requires llama.cpp for MTP
- No vision support
- License is apache‑2.0
Best for
- ARC‑Challenge style reasoning
- Tool‑augmented workflows
- Fast generation with self‑speculative decoding
Who it’s for. Developers building reasoning‑heavy applications who need a large‑context model and can run llama.cpp. Suitable for research and production use where the apache‑2.0 license is acceptable.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ1_M | 7.2 GB | 8.4 GB | Too big | Get file → | |
| IQ2_XXS | 8.2 GB | 9.4 GB | Too big | Get file → | |
| IQ2_M | 10.0 GB | 11.1 GB | Too big | Get file → | |
| Q2_K | 10.8 GB | 12.0 GB | Too big | Get file → | |
| IQ3_XXS | 11.5 GB | 12.7 GB | Too big | Get file → | |
| IQ3_M | 12.9 GB | 14.1 GB | Too big | Get file → | |
| Q3_K_M | 14.0 GB | 15.2 GB | Too big | Get file → | |
| IQ4_XS | 15.7 GB | 16.9 GB | Too big | Get file → | |
| Q4_K_S | 16.6 GB | 17.8 GB | Too big | Get file → | |
| Q4_K_M | 17.7 GB | 18.9 GB | Too big | Get file → | |
| Q5_K_M | 20.7 GB | 21.8 GB | Too big | Get file → | |
| Q6_K | 23.8 GB | 25.0 GB | Too big | Get file → | |
| Q8_0 | 30.8 GB | 32.0 GB | Too big | Get file → | |
| BF16 | 57.9 GB | 59.1 GB | Too big | Get file → |
For 8 GB VRAM, use IQ1_M (7.19 GB, ~8.4 GB memory at 8 k context). For 12 GB VRAM, IQ2_M (9.96 GB, ~11.1 GB memory). For 16 GB VRAM, Q4_K_M (17.71 GB, ~18.9 GB memory).
What GPU do you need for Qwen3.6 29B REAP Opus Reasoning Distill MTP?
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 (10.8 GB file, about 12.0 GB in use).
- 16 GB VRAM: the best quant that fits is Q3_K_M (14.0 GB file, about 15.2 GB in use).
- 24 GB VRAM: the best quant that fits is Q5_K_M (20.7 GB file, about 21.8 GB in use).
- 32 GB VRAM: the best quant that fits is Q8_0 (30.8 GB file, about 32.0 GB in use).
- 48 GB VRAM: the best quant that fits is Q8_0 (30.8 GB file, about 32.0 GB in use).
Which Qwen3.6 29B REAP Opus Reasoning Distill MTP quant should you choose?
Q4_K_M is the usual starting point for Qwen3.6 29B REAP Opus Reasoning Distill MTP: it balances file size and quality well. Here it is a 17.7 GB file.
The Q8_0 file is 30.8 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 IQ1_M at 7.2 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
Frequently asked questions about Qwen3.6 29B REAP Opus Reasoning Distill MTP
How much VRAM does Qwen3.6 29B REAP Opus Reasoning Distill MTP need?
About 18.9 GB at Q4_K_M with an 8K context window. That is the 17.7 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 29B REAP Opus Reasoning Distill MTP on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 8.4 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Qwen3.6 29B REAP Opus Reasoning Distill MTP GGUF?
It depends on the quant: IQ1_M is 7.2 GB, IQ2_XXS is 8.2 GB, IQ2_M is 10.0 GB, Q2_K is 10.8 GB, IQ3_XXS is 11.5 GB, IQ3_M is 12.9 GB and 8 more.
What context length does Qwen3.6 29B REAP Opus Reasoning Distill MTP 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 2.8 GB at 32K.
Where can I download Qwen3.6 29B REAP Opus Reasoning Distill MTP GGUF?
The files are on Hugging Face in the barozp/Qwen3.6-29B-REAP-Opus-Reasoning-Distill-MTP-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 29B REAP Opus Reasoning Distill MTP 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?
Yes, using the IQ1_M quantization which is 7.19 GB and requires about 8.4 GB memory at 8 k context.
Does this model support vision?
No, the model has no vision capabilities.
Is the MTP head available in transformers?
No, transformers does not implement MTP inference; it is only usable via llama.cpp's self‑speculative decoding.