LLM EXPLORER 65,695 MODELS INDEXED

MiniMax M3.1 Flash EXL3 EAQ V1.3.06 Bpw by Mia-AiLab

By Mia-AiLab · 165 downloads
HF Repo ↗ Copy URL

MiniMax M3.1 Flash EXL3 EAQ V1.3.06 Bpw is an open-source language model by Mia-AiLab. Features: 84.3b LLM, VRAM: 169.4GB.

Base model:minimaxai/minimax-m... Base model:quantized:minimaxai...   Conversational   Custom code   Eaq   Exl3   Exllamav3   Minimax   Minimax m3 vl   Moe   Quantized   Region:us   Safetensors   Sharded   Tensorflow   Vision
Model Card on HF 🤗: https://huggingface.co/Mia-AiLab/MiniMax-M3.1-Flash-EXL3-EAQ-v1-3.06-bpw 

MiniMax M3.1 Flash EXL3 EAQ V1.3.06 Bpw Parameters and Internals

LLM NameMiniMax M3.1 Flash EXL3 EAQ V1.3.06 Bpw
Repository 🤗https://huggingface.co/Mia-AiLab/MiniMax-M3.1-Flash-EXL3-EAQ-v1-3.06-bpw 
Base Model(s)  MiniMaxAI/MiniMax-M3.1-Flash   MiniMaxAI/MiniMax-M3.1-Flash
Model Size84.3b
Required VRAM169.4 GB
Updated2026-10-27
MaintainerMia-AiLab
Model Typeminimax_m3_vl
Model Files  6.0 GB: 1-of-20   8.6 GB: 2-of-20   8.6 GB: 3-of-20   8.6 GB: 4-of-20   8.6 GB: 5-of-20   8.6 GB: 6-of-20   8.6 GB: 7-of-20   8.6 GB: 8-of-20   8.6 GB: 9-of-20   8.6 GB: 10-of-20   8.6 GB: 11-of-20   8.6 GB: 12-of-20   8.6 GB: 13-of-20   8.6 GB: 14-of-20   8.6 GB: 15-of-20   8.6 GB: 16-of-20   8.6 GB: 17-of-20   8.6 GB: 18-of-20   8.6 GB: 19-of-20   6.9 GB: 20-of-20   1.7 GB: 1-of-1
Model ArchitectureMiniMaxM3SparseForConditionalGeneration
Model Max Length40960000
Transformers Version4.52.4
Tokenizer ClassPreTrainedTokenizerFast
Padding Token]!p~[
Torch Data Typebfloat16