LLM EXPLORER 60,062 MODELS INDEXED

MiniMax M2.7 REAP 172B A10B AutoRound W4A16 by MJPansa

By MJPansa · 3060 downloads

MiniMax M2.7 REAP 172B A10B AutoRound W4A16 is an open-source language model by MJPansa. Features: 172b LLM, VRAM: 91.4GB, Context: 192K, License: other, LLM Explorer Score: 0.28.

  4-bit   Auto-round   Autoround Base model:minimaxai/minimax-m... Base model:quantized:minimaxai...   Blackwell   Conversational   Custom code Dataset:theblackcat102/evol-co...   Dgx-spark   En   Endpoints compatible   Gptq   Int4   Minimax   Minimax-m2   Minimax m2   Mixture-of-experts   Moe   Pruned   Quantization   Reap   Region:us   Safetensors   Sharded   Tensorflow   Vllm   W4a16

MiniMax M2.7 REAP 172B A10B AutoRound W4A16 Parameters and Internals

LLM NameMiniMax M2.7 REAP 172B A10B AutoRound W4A16
Repository πŸ€—https://huggingface.co/MJPansa/MiniMax-M2.7-REAP-172B-A10B-AutoRound-W4A16 
Base Model(s)  MiniMaxAI/MiniMax-M2.7   MiniMaxAI/MiniMax-M2.7
Model Size172b
Required VRAM91.4 GB
Updated2026-07-26
MaintainerMJPansa
Model Typeminimax_m2
Model Files  4.0 GB: 1-of-23   4.1 GB: 2-of-23   4.0 GB: 3-of-23   4.1 GB: 4-of-23   4.0 GB: 5-of-23   4.1 GB: 6-of-23   4.0 GB: 7-of-23   4.1 GB: 8-of-23   4.0 GB: 9-of-23   4.0 GB: 10-of-23   4.0 GB: 11-of-23   4.0 GB: 12-of-23   4.1 GB: 13-of-23   4.1 GB: 14-of-23   4.0 GB: 15-of-23   4.0 GB: 16-of-23   4.0 GB: 17-of-23   4.1 GB: 18-of-23   4.0 GB: 19-of-23   4.1 GB: 20-of-23   4.0 GB: 21-of-23   4.0 GB: 22-of-23   2.6 GB: 23-of-23
Supported Languagesen
Model ArchitectureMiniMaxM2ForCausalLM
Licenseother
Context Length196608
Model Max Length196608
Transformers Version4.55.0
Tokenizer ClassGPT2Tokenizer
Vocabulary Size200064
Torch Data Typefloat32

Best Alternatives to MiniMax M2.7 REAP 172B A10B AutoRound W4A16

Best Alternatives
Context / RAM
Downloads
Likes
...M2.7 REAP 172B A10B NVFP4 GB10192K / 98.9 GB175036
...niMax M2.7 REAP 172B A10B BF16192K / 205 GB181