Download MiniMax-01 Models

Choose between MiniMax-Text-01 and MiniMax-VL-01 models

MiniMax-Text-01

Advanced language model with hybrid attention and MoE architecture

Text
  • 456B total parameters
  • 45.9B activated parameters
  • 4M token context length
  • 80-layer architecture
Download Text Model

MiniMax-VL-01

Vision-language model built on MiniMax-Text-01

Vision-Language
  • 303M ViT parameters
  • Dynamic resolution
  • 336×336 to 2016×2016
  • Advanced visual processing
Download VL Model

Installation Instructions

Access models through Hugging Face:

# For Text Model
git lfs install
git clone https://huggingface.co/MiniMaxAI/MiniMax-Text-01

# For VL Model
git lfs install
git clone https://huggingface.co/MiniMaxAI/MiniMax-VL-01