MiniMax-01 series
Open models MiniMax-Text-01 and MiniMax-VL-01: 456B parameters, 45.9B active, with up to 4M tokens of context.
MiniMax launches open-source MiniMax-01 series with Text-01 & VL-01 models. Lightning Attention architecture, 456B parameters, 4M token context, 20-32x longer than competitors. Starting $0.2/M tokens.
Official source
MiniMax-01 is Now Open-Source: Scaling Lightning Attention for the AI Agent Eraminimax.io · published