MiniMax M2.1 REAP 40 needs about 86.8 GB of memory at Q4_K_M (a 84.2 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 24 quants, from 0.3 GB to 114.3 GB (Q6_K). Longer context needs more: at 32K tokens the KV cache alone adds about 8.3 GB.
MiniMax M2.1 REAP 40 – 139B parameter model pruned for efficiency
MiniMax M2.1 REAP 40 is a 139B parameter model pruned by 40% using REAP. It supports reasoning and tools. The model is released under apache-2.0 and offers a large context window of 196608 tokens.
Good at
- Reasoning
- Tool usage
- Large context handling
Limits
- No vision support
- Requires large VRAM for full precision
- License is apache-2.0
Best for
- Complex reasoning tasks
- Code generation with tool calls
- Long‑form content creation
Who it’s for. Developers needing a high‑capacity language model with expert pruning. Researchers working on large‑scale inference and those who can manage the memory requirements.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 0.3 GB | 2.9 GB | Fits | Get file → | |
| IQ1_S | 28.5 GB | 31.1 GB | Too big | Get file → | |
| IQ1_M | 31.6 GB | 34.2 GB | Too big | Get file → | |
| IQ2_XXS | 36.7 GB | 39.3 GB | Too big | Get file → | |
| IQ2_XS | 40.9 GB | 43.4 GB | Too big | Get file → | |
| IQ2_S | 41.6 GB | 44.2 GB | Too big | Get file → | |
| IQ2_M | 45.7 GB | 48.3 GB | Too big | Get file → | |
| Q2_K_S | 47.3 GB | 49.9 GB | Too big | Get file → | |
| Q2_K | 50.8 GB | 53.4 GB | Too big | Get file → | |
| IQ3_XXS | 53.6 GB | 56.2 GB | Too big | Get file → | |
| IQ3_XS | 56.9 GB | 59.5 GB | Too big | Get file → | |
| Q3_K_S | 60.1 GB | 62.7 GB | Too big | Get file → | |
| IQ3_S | 60.2 GB | 62.8 GB | Too big | Get file → | |
| IQ3_M | 61.0 GB | 63.6 GB | Too big | Get file → | |
| Q3_K_M | 66.6 GB | 69.2 GB | Too big | Get file → | |
| Q3_K_L | 72.1 GB | 74.7 GB | Too big | Get file → | |
| IQ4_XS | 74.2 GB | 76.8 GB | Too big | Get file → | |
| Q4_0 | 78.9 GB | 81.4 GB | Too big | Get file → | |
| Q4_K_S | 79.2 GB | 81.8 GB | Too big | Get file → | |
| Q4_K_M | 84.2 GB | 86.8 GB | Too big | Get file → | |
| Q4_1 | 87.2 GB | 89.8 GB | Too big | Get file → | |
| Q5_K_S | 95.9 GB | 98.4 GB | Too big | Get file → | |
| Q5_K_M | 98.8 GB | 101.3 GB | Too big | Get file → | |
| Q6_K | 114.3 GB | 116.8 GB | Too big | Get file → |
For 8 GB VRAM use the "unknown" quant (0.3 GB file, ~2.9 GB memory at 8 k context). For 16 GB VRAM the IQ1_S quant (28.51 GB file, ~31.1 GB memory) is a good balance. For 24 GB VRAM the IQ1_M quant (31.59 GB file, ~34.2 GB memory) offers higher quality.
What GPU do you need for MiniMax M2.1 REAP 40?
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: the best quant that fits is the main file (0.3 GB file, about 2.9 GB in use).
- 6 GB VRAM: the best quant that fits is the main file (0.3 GB file, about 2.9 GB in use).
- 8 GB VRAM: the best quant that fits is the main file (0.3 GB file, about 2.9 GB in use).
- 12 GB VRAM: the best quant that fits is the main file (0.3 GB file, about 2.9 GB in use).
- 16 GB VRAM: the best quant that fits is the main file (0.3 GB file, about 2.9 GB in use).
- 24 GB VRAM: the best quant that fits is the main file (0.3 GB file, about 2.9 GB in use).
- 32 GB VRAM: the best quant that fits is IQ1_S (28.5 GB file, about 31.1 GB in use).
- 48 GB VRAM: the best quant that fits is IQ2_S (41.6 GB file, about 44.2 GB in use).
Which MiniMax M2.1 REAP 40 quant should you choose?
Q4_K_M is the usual starting point for MiniMax M2.1 REAP 40: it balances file size and quality well. Here it is a 84.2 GB file.
If memory is tight, the smallest file is the first one in the table at 0.3 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 MiniMax M2.1 REAP 40
How much VRAM does MiniMax M2.1 REAP 40 need?
About 86.8 GB at Q4_K_M with an 8K context window. That is the 84.2 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 MiniMax M2.1 REAP 40 on an 8 GB GPU?
Yes, with the smaller files: the 0.3 GB file needs about 2.9 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of MiniMax M2.1 REAP 40 GGUF?
It depends on the quant: file is 0.3 GB, IQ1_S is 28.5 GB, IQ1_M is 31.6 GB, IQ2_XXS is 36.7 GB, IQ2_XS is 40.9 GB, IQ2_S is 41.6 GB and 18 more.
What context length does MiniMax M2.1 REAP 40 support?
The GGUF header lists a trained context length of 196,608 tokens (192K). The KV cache grows with the context you use: about 2.1 GB at 8K and 8.3 GB at 32K.
Where can I download MiniMax M2.1 REAP 40 GGUF?
The files are on Hugging Face in the mradermacher/MiniMax-M2.1-REAP-40-i1-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 MiniMax M2.1 REAP 40 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 MiniMax M2.1 REAP 40 on 8 GB VRAM?
Yes, using the "unknown" quant (0.3 GB file) which requires about 2.9 GB memory at 8 k context.
What is REAP pruning?
REAP is Router‑weighted Expert Activation Pruning, a method that removes 40% of experts based on router activation patterns.
Under what license is the model released?
The model is released under the apache-2.0 license.