LLM EXPLORER 62,939 MODELS INDEXED

MiniMax M3 MXFP4 AttnFP8 by amd

By amd · 129 downloads
HF Repo ↗ Copy URL

MiniMax M3 MXFP4 AttnFP8 is an open-source language model by amd. Features: 231.9b LLM, VRAM: 164.5GB, License: other, LLM Explorer Score: 0.24.

  8-bit   Agent Base model:minimaxai/minimax-m... Base model:quantized:minimaxai...   Coding   Conversational   Custom code   Endpoints compatible   Image-text-to-text   Minimax m3 vl   Moe   Multimodal   Quark   Region:us   Safetensors   Sharded   Tensorflow   Video
Model Card on HF 🤗: https://huggingface.co/amd/MiniMax-M3-MXFP4-AttnFP8 

MiniMax M3 MXFP4 AttnFP8 Parameters and Internals

LLM NameMiniMax M3 MXFP4 AttnFP8
Repository 🤗https://huggingface.co/amd/MiniMax-M3-MXFP4-AttnFP8 
Base Model(s)  MiniMax M3   MiniMaxAI/MiniMax-M3
Model Size231.9b
Required VRAM164.5 GB
Updated2026-08-09
Maintaineramd
Model Typeminimax_m3_vl
Model Files  5.5 GB: 1-of-59   3.7 GB: 2-of-59   4.3 GB: 3-of-59   4.3 GB: 4-of-59   4.3 GB: 5-of-59   4.3 GB: 6-of-59   4.3 GB: 7-of-59   4.3 GB: 8-of-59   4.3 GB: 9-of-59   4.3 GB: 10-of-59   4.3 GB: 11-of-59   4.3 GB: 12-of-59   4.3 GB: 13-of-59   4.3 GB: 14-of-59   4.3 GB: 15-of-59   4.3 GB: 16-of-59   4.3 GB: 17-of-59   4.3 GB: 18-of-59   4.3 GB: 19-of-59   4.3 GB: 20-of-59   4.3 GB: 21-of-59   4.3 GB: 22-of-59   4.3 GB: 23-of-59   4.3 GB: 24-of-59   1.4 GB: 25-of-59   4.5 GB: 26-of-59   4.0 GB: 27-of-59   4.0 GB: 28-of-59   4.0 GB: 29-of-59   4.0 GB: 30-of-59   4.0 GB: 31-of-59   4.0 GB: 32-of-59   4.0 GB: 33-of-59   4.0 GB: 34-of-59   4.0 GB: 35-of-59   4.0 GB: 36-of-59   3.7 GB: 37-of-59   3.7 GB: 38-of-59   3.7 GB: 39-of-59   3.7 GB: 40-of-59
Model ArchitectureMiniMaxM3SparseForConditionalGeneration
Licenseother
Transformers Version4.52.4
Torch Data Typebfloat16

Best Alternatives to MiniMax M3 MXFP4 AttnFP8

Best Alternatives
Context / RAM
Downloads
Likes
MiniMax M3 MXFP40K / 168.7 GB1131444
Note: green Score (e.g. "73.2") means that the model is better than amd/MiniMax-M3-MXFP4-AttnFP8.