What are the hardware requirements for MiniMax M2.7 REAP 172B A10B NVFP4 GB10?

MiniMax M2.7 REAP 172B A10B NVFP4 GB10 requires approximately 98.9 GB of VRAM and supports a context window of 192K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed MiniMax M2.7 REAP 172B A10B NVFP4 GB10 and how large is it?

MiniMax M2.7 REAP 172B A10B NVFP4 GB10 is developed by saricles, a model with 172b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate MiniMax M2.7 REAP 172B A10B NVFP4 GB10?

MiniMax M2.7 REAP 172B A10B NVFP4 GB10 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

MiniMax M2.7 REAP 172B A10B NVFP4 GB10 by saricles — VRAM 98.9GB, 192K context

Name: MiniMax M2.7 REAP 172B A10B NVFP4 GB10
Author: saricles

MiniMax M2.7 REAP 172B A10B NVFP4 GB10 is an open-source language model by saricles. Features: 172b LLM, VRAM: 98.9GB, Context: 192K, License: other, LLM Explorer Score: 0.28.

Arxiv:2510.13999 4-bit 8-bit Agentic Base model:finetune:saricles/m... Base model:saricles/minimax-m2... Compressed-tensors Conversational Custom code Dgx-spark En Endpoints compatible Gb10 Minimax Minimax m2 Moe Nvfp4 Quantized Reap Region:us Safetensors Sharded Tensorflow Tool-use Vllm Zh

Model Card on HF 🤗: https://huggingface.co/saricles/MiniMax-M2.7-REAP-172B-A10B-NVFP4-GB10

MiniMax M2.7 REAP 172B A10B NVFP4 GB10 Parameters and Internals

LLM Name	MiniMax M2.7 REAP 172B A10B NVFP4 GB10
Repository 🤗	https://huggingface.co/saricles/MiniMax-M2.7-REAP-172B-A10B-NVFP4-GB10
Base Model(s)	saricles/MiniMax-M2.7-REAP-172B-A10B-BF16 saricles/MiniMax-M2.7-REAP-172B-A10B-BF16
Model Size	172b
Required VRAM	98.9 GB
Updated	2026-08-07
Maintainer	saricles
Model Type	minimax_m2
Model Files	5.0 GB: 1-of-20 5.0 GB: 2-of-20 5.0 GB: 3-of-20 5.0 GB: 4-of-20 5.0 GB: 5-of-20 5.0 GB: 6-of-20 5.0 GB: 7-of-20 5.0 GB: 8-of-20 5.0 GB: 9-of-20 5.0 GB: 10-of-20 5.0 GB: 11-of-20 5.0 GB: 12-of-20 5.0 GB: 13-of-20 5.0 GB: 14-of-20 5.0 GB: 15-of-20 5.0 GB: 16-of-20 5.0 GB: 17-of-20 5.0 GB: 18-of-20 5.0 GB: 19-of-20 3.9 GB: 20-of-20
Supported Languages	en zh
Model Architecture	MiniMaxM2ForCausalLM
License	other
Context Length	196608
Model Max Length	196608
Transformers Version	4.57.6
Tokenizer Class	GPT2Tokenizer
Vocabulary Size	200064

Rank the MiniMax M2.7 REAP 172B A10B NVFP4 GB10 capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

Best Alternatives to MiniMax M2.7 REAP 172B A10B NVFP4 GB10

Best Alternatives	Context / RAM	Downloads	Likes
...REAP 172B A10B AutoRound W4A16	192K / 91.4 GB	3060	11
...niMax M2.7 REAP 172B A10B BF16	192K / 205 GB	18	1

Note: green Score (e.g. "73.2") means that the model is better than saricles/MiniMax-M2.7-REAP-172B-A10B-NVFP4-GB10.