MI

cerebras/MiniMax-M2-REAP-162B-A10B

MinimaxGGUF by bartowskiUpdated 2025-11-19
Download on Hugging Face
Parameters162B
Context192K
Layers62
Architectureminimax-m2
Smallest quant0.3 GB
Quick answer

cerebras/MiniMax-M2-REAP-162B-A10B needs about 100.8 GB of memory at Q4_K_M (a 98.2 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 26 quants, from 0.3 GB to 172.2 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 8.3 GB.

MiniMax-M2-REAP-162B-A10B – a 30% lighter, sparse MoE model

MiniMax-M2-REAP-162B-A10B is a compressed variant of MiniMax-M2, reduced from 230B to 162B parameters while keeping near‑identical performance. It uses REAP pruning to keep 10B active parameters per token and supports a 196,608‑token context length.

Good at

  • Reasoning
  • Tools usage
  • Code generation

Limits

  • Requires large VRAM for higher quants
  • No vision support
  • License is other

Best for

  • Local deployment on limited GPU
  • Academic research
  • Resource‑constrained inference

Who it’s for. Developers and researchers needing a lightweight, high‑capacity language model that can run on local hardware. The model is suitable for applications that require reasoning, tool calling, and code generation without the need for vision capabilities.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—0.3 GB 2.9 GB
Fits Get file →
IQ1_S33.6 GB 36.2 GB
Too big Get file →
IQ1_M35.0 GB 37.6 GB
Too big Get file →
IQ2_XXS39.0 GB 41.6 GB
Too big Get file →
IQ2_XS44.9 GB 47.5 GB
Too big Get file →
IQ2_S45.2 GB 47.7 GB
Too big Get file →
IQ2_M51.2 GB (2 parts) 53.8 GB
Too big Get file →
Q2_K56.8 GB (2 parts) 59.4 GB
Too big Get file →
Q2_K_L57.4 GB (2 parts) 60.0 GB
Too big Get file →
IQ3_XXS64.0 GB (2 parts) 66.6 GB
Too big Get file →
IQ3_XS66.6 GB (2 parts) 69.2 GB
Too big Get file →
Q3_K_S70.4 GB (2 parts) 73.0 GB
Too big Get file →
IQ3_M73.8 GB (2 parts) 76.4 GB
Too big Get file →
Q3_K_M73.8 GB (2 parts) 76.4 GB
Too big Get file →
Q3_K_L76.7 GB (2 parts) 79.3 GB
Too big Get file →
Q3_K_XL77.3 GB (2 parts) 79.8 GB
Too big Get file →
IQ4_XS86.7 GB (3 parts) 89.3 GB
Too big Get file →
IQ4_NL91.7 GB (3 parts) 94.2 GB
Too big Get file →
Q4_093.1 GB (3 parts) 95.7 GB
Too big Get file →
Q4_K_S94.8 GB (3 parts) 97.4 GB
Too big Get file →
Q4_K_M98.2 GB (3 parts) 100.8 GB
Too big Get file →
Q4_1101.6 GB (3 parts) 104.1 GB
Too big Get file →
Q5_K_S111.6 GB (3 parts) 114.2 GB
Too big Get file →
Q5_K_M115.0 GB (3 parts) 117.6 GB
Too big Get file →
Q6_K133.0 GB (4 parts) 135.6 GB
Too big Get file →
Q8_0172.2 GB (5 parts) 174.8 GB
Too big Get file →

For 8 GB VRAM, use IQ1_S or IQ1_M quants (file size 33.57 GB or 34.99 GB). For 12 GB VRAM, Q2_XXS (39.02 GB) or Q3_K_M (73.81 GB) are options. Higher quants such as Q4_K_M (98.23 GB) or Q5_K_M (115.03 GB) need more memory; split shards can reduce per‑GPU load.

What GPU do you need for cerebras/MiniMax-M2-REAP-162B-A10B?

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 the main file (0.3 GB file, about 2.9 GB in use).
  • 48 GB VRAM: the best quant that fits is IQ2_S (45.2 GB file, about 47.7 GB in use).

Which cerebras/MiniMax-M2-REAP-162B-A10B quant should you choose?

Q4_K_M is the usual starting point for cerebras/MiniMax-M2-REAP-162B-A10B: it balances file size and quality well. Here it is a 98.2 GB file.

The Q8_0 file is 172.2 GB, about 1.8 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 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

Architectureminimax-m2
Parameters162B (approximate, mixture of experts)
Active parameters10B
Layers62
Trained context196,608 tokens
KV cache (fp16)248 KB per token

Source

Quantized bybartowski
FamilyMinimax
Licenseother
Downloads (HF)685
Last updated2025-11-19

Frequently asked questions about cerebras/MiniMax-M2-REAP-162B-A10B

How much VRAM does cerebras/MiniMax-M2-REAP-162B-A10B need?

About 100.8 GB at Q4_K_M with an 8K context window. That is the 98.2 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 cerebras/MiniMax-M2-REAP-162B-A10B 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 cerebras/MiniMax-M2-REAP-162B-A10B GGUF?

It depends on the quant: file is 0.3 GB, IQ1_S is 33.6 GB, IQ1_M is 35.0 GB, IQ2_XXS is 39.0 GB, IQ2_XS is 44.9 GB, IQ2_S is 45.2 GB and 20 more.

What context length does cerebras/MiniMax-M2-REAP-162B-A10B 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 cerebras/MiniMax-M2-REAP-162B-A10B GGUF?

The files are on Hugging Face in the bartowski/cerebras_MiniMax-M2-REAP-162B-A10B-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 cerebras/MiniMax-M2-REAP-162B-A10B 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 or IQ1_M quant, which estimate 36.2 GB or 37.6 GB memory at 8k context, but actual VRAM usage will be lower due to batching.

Does this model support vision tasks?

No, the model has no vision capabilities.

What is the context length of this model?

The model supports a context length of 196,608 tokens.

Other GGUFs of MiniMax-M2-REAP-162B-A10B

Similar models

Browse all →

More from bartowski

All by bartowski →

More Minimax models

All Minimax →

Other large models

All large models →