MiniMax M2.5 REAP 172B A10B i1 needs about 106.9 GB of memory at Q4_K_M (a 104.4 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 24 quants, from 0.4 GB to 141.6 GB (Q6_K). Longer context needs more: at 32K tokens the KV cache alone adds about 8.3 GB.
MiniMax M2.5 REAP 172B A10B i1 – a 172B parameter sparse MoE model
MiniMax M2.5 REAP 172B A10B i1 is a 172B parameter sparse Mixture-of-Experts model with 10B active parameters per token. It supports a context length of 196,608 tokens and is compressed by 25% using REAP. The model is available in multiple GGUF quantizations.
Good at
- Reasoning
- Tool calling
- Long‑context handling
Limits
- No vision support
- Requires large VRAM for higher quants
- License is other
Best for
- Academic research on sparse MoE
- Local deployments with limited GPU memory
- Testing long‑context generation
Who it’s for. Researchers and developers who need a memory‑efficient variant of MiniMax‑M2.5 for local inference. The model is compatible with vanilla vLLM and does not require source modifications.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 0.4 GB | 3.0 GB | Fits | Get file → | |
| IQ1_S | 35.2 GB | 37.8 GB | Too big | Get file → | |
| IQ1_M | 39.0 GB | 41.6 GB | Too big | Get file → | |
| IQ2_XXS | 45.4 GB | 48.0 GB | Too big | Get file → | |
| IQ2_XS | 50.6 GB | 53.2 GB | Too big | Get file → | |
| IQ2_S | 51.5 GB | 54.0 GB | Too big | Get file → | |
| IQ2_M | 56.6 GB | 59.1 GB | Too big | Get file → | |
| Q2_K_S | 58.6 GB | 61.2 GB | Too big | Get file → | |
| Q2_K | 62.9 GB | 65.5 GB | Too big | Get file → | |
| IQ3_XXS | 66.4 GB | 69.0 GB | Too big | Get file → | |
| IQ3_XS | 70.5 GB | 73.1 GB | Too big | Get file → | |
| Q3_K_S | 74.5 GB | 77.1 GB | Too big | Get file → | |
| IQ3_S | 74.5 GB | 77.1 GB | Too big | Get file → | |
| IQ3_M | 75.5 GB | 78.1 GB | Too big | Get file → | |
| Q3_K_M | 82.5 GB | 85.1 GB | Too big | Get file → | |
| Q3_K_L | 89.3 GB | 91.9 GB | Too big | Get file → | |
| IQ4_XS | 92.0 GB | 94.6 GB | Too big | Get file → | |
| Q4_0 | 97.7 GB | 100.3 GB | Too big | Get file → | |
| Q4_K_S | 98.1 GB | 100.7 GB | Too big | Get file → | |
| Q4_K_M | 104.4 GB | 106.9 GB | Too big | Get file → | |
| Q4_1 | 108.1 GB | 110.7 GB | Too big | Get file → | |
| Q5_K_S | 118.8 GB | 121.4 GB | Too big | Get file → | |
| Q5_K_M | 122.4 GB | 125.0 GB | Too big | Get file → | |
| Q6_K | 141.6 GB | 144.2 GB | Too big | Get file → |
For 8 GB VRAM, use IQ1_S (file size 35.21 GB, memory 37.8 GB at 8 k context). For 12 GB VRAM, IQ2_XXS (45.44 GB, 48 GB). For 24 GB VRAM, Q4_K_M (104.36 GB, 106.9 GB). Estimates are approximate.
What GPU do you need for MiniMax M2.5 REAP 172B A10B i1?
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 (0.4 GB file, about 3.0 GB in use).
- 6 GB VRAM: the best quant that fits is the main file (0.4 GB file, about 3.0 GB in use).
- 8 GB VRAM: the best quant that fits is the main file (0.4 GB file, about 3.0 GB in use).
- 12 GB VRAM: the best quant that fits is the main file (0.4 GB file, about 3.0 GB in use).
- 16 GB VRAM: the best quant that fits is the main file (0.4 GB file, about 3.0 GB in use).
- 24 GB VRAM: the best quant that fits is the main file (0.4 GB file, about 3.0 GB in use).
- 32 GB VRAM: the best quant that fits is the main file (0.4 GB file, about 3.0 GB in use).
- 48 GB VRAM: the best quant that fits is IQ2_XXS (45.4 GB file, about 48.0 GB in use).
Which MiniMax M2.5 REAP 172B A10B i1 quant should you choose?
Q4_K_M is the usual starting point for MiniMax M2.5 REAP 172B A10B i1: it balances file size and quality well. Here it is a 104.4 GB file.
If memory is tight, the smallest file is the first one in the table at 0.4 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 M2.5 REAP 172B A10B i1
How much VRAM does MiniMax M2.5 REAP 172B A10B i1 need?
About 106.9 GB at Q4_K_M with an 8K context window. That is the 104.4 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 MiniMax M2.5 REAP 172B A10B i1 on an 8 GB GPU?
Yes, with the smaller files: the 0.4 GB file needs about 3.0 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of MiniMax M2.5 REAP 172B A10B i1 GGUF?
It depends on the quant: file is 0.4 GB, IQ1_S is 35.2 GB, IQ1_M is 39.0 GB, IQ2_XXS is 45.4 GB, IQ2_XS is 50.6 GB, IQ2_S is 51.5 GB and 18 more.
What context length does MiniMax M2.5 REAP 172B A10B i1 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 M2.5 REAP 172B A10B i1 GGUF?
The files are on Hugging Face in the mradermacher/MiniMax-M2.5-REAP-172B-A10B-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 M2.5 REAP 172B A10B i1 use?
Hugging Face lists the license as other. 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?
Use the IQ1_S quant, which requires about 37.8 GB of VRAM at 8 k context.
Does this model support vision tasks?
No, the model has no vision capabilities.
What is the context length of this model?
The model supports up to 196,608 tokens.