LA

MiniMax M2.7 REAP 139B A10B

LabGGUF by dervigUpdated 2026-04-16
Download on Hugging Face
Parameters139B
Context192K
Layers62
Architectureminimax-m2
Smallest quant66.6 GB
Quick answer

MiniMax M2.7 REAP 139B A10B 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 6 quants, from 66.6 GB (Q3_K_M) to 147.9 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 8.3 GB.

MiniMax M2.7 REAP 139B A10B – 139 B pruned MoE LLM

MiniMax M2.7 REAP 139B A10B is a 139 B parameter Mixture‑of‑Experts model pruned 40 % with REAP. It retains 10 B active parameters per token and is quantized for GGUF inference. The model supports reasoning and tool usage.

Good at

  • Reasoning
  • Tool integration
  • Large context (196 608 tokens)

Limits

  • No vision support
  • Requires high VRAM for full quality
  • License is other

Best for

  • Code generation
  • Complex reasoning tasks
  • Tool‑enabled workflows

Who it’s for. Developers needing a large‑scale reasoning model that fits on high‑end consumer GPUs. Ideal for users with 96 GB Apple Silicon or equivalent VRAM who want near‑baseline performance with a pruned MoE architecture.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q3_K_M66.6 GB 69.2 GB
Too big Get file →
IQ4_XS74.2 GB 76.8 GB
Too big Get file →
IQ4_NL80.1 GB 82.6 GB
Too big Get file →
Q4_K_M84.2 GB 86.8 GB
Too big Get file →
Q6_K114.3 GB 116.8 GB
Too big Get file →
Q8_0147.9 GB 150.5 GB
Too big Get file →

For 96 GB Apple Silicon, Q4_K_M (~84 GB) is recommended. With 64 GB, Q3_K_M (~66 GB) is the only variant that fits, though reasoning quality may drop. For 128 GB, Q6_K (~114 GB) offers near‑baseline quality. Larger systems can use Q8_0 (~148 GB).

What GPU do you need for MiniMax M2.7 REAP 139B 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: 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: no quant fits at an 8K context.
  • 16 GB VRAM: no quant fits at an 8K context.
  • 24 GB VRAM: no quant fits at an 8K context.
  • 32 GB VRAM: no quant fits at an 8K context.
  • 48 GB VRAM: no quant fits at an 8K context.

Which MiniMax M2.7 REAP 139B A10B quant should you choose?

Q4_K_M is the usual starting point for MiniMax M2.7 REAP 139B A10B: it balances file size and quality well. Here it is a 84.2 GB file.

The Q8_0 file is 147.9 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 Q3_K_M at 66.6 GB, with the quality trade-offs that come with fewer bits. A reasonable choice when a 4-bit file will not fit, mostly on larger models. Quality loss is noticeable but often acceptable.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureminimax-m2
Parameters139B (approximate, mixture of experts)
Active parameters10B
Layers62
Trained context196,608 tokens
KV cache (fp16)248 KB per token

Source

Quantized bydervig
FamilyLab
Licenseother
Downloads (HF)294
Last updated2026-04-16

Frequently asked questions about MiniMax M2.7 REAP 139B A10B

How much VRAM does MiniMax M2.7 REAP 139B A10B 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 M2.7 REAP 139B A10B on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 69.2 GB, so you would need a larger GPU or CPU offloading.

What is the file size of MiniMax M2.7 REAP 139B A10B GGUF?

It depends on the quant: Q3_K_M is 66.6 GB, IQ4_XS is 74.2 GB, IQ4_NL is 80.1 GB, Q4_K_M is 84.2 GB, Q6_K is 114.3 GB, Q8_0 is 147.9 GB.

What context length does MiniMax M2.7 REAP 139B 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 MiniMax M2.7 REAP 139B A10B GGUF?

The files are on Hugging Face in the dervig/m51Lab-MiniMax-M2.7-REAP-139B-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 MiniMax M2.7 REAP 139B 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?

No, the smallest quantization is Q3_K_M at 66.58 GB, which exceeds 8 GB VRAM.

Does the model support vision tasks?

No, the model has no vision capabilities.

What context length does it support?

It supports up to 196 608 tokens in training, but inference is limited by VRAM and quantization.

Other GGUFs of m51Lab-MiniMax-M2.7-REAP-139B-A10B

Similar models

Browse all →

More from dervig

All by dervig →

More Lab models

All Lab →

Other large models

All large models →