MiniMax-M2.1
Lightweight coding-focused model with strong multi-language programming and agentic capabilities
MiniMax-M2.1
MiniMax • December 2025
Training Data
Up to late 2025
MiniMax-M2.1
December 2025
Parameters
230B (10B active)
Training Method
Mixture of Experts
Context Window
196,000 tokens
Knowledge Cutoff
November 2025
Key Features
Multi-language Coding • Agentic Workflows • Open Weights • Cost Efficient
Capabilities
Coding: Excellent (74% SWE-bench)
Efficiency: Outstanding
Multilingual: Excellent
What's New in This Version
Enhanced multi-language programming with open weights at $0.27/$1.12 per M tokens
Lightweight coding-focused model with strong multi-language programming and agentic capabilities
What's New in This Version
Enhanced multi-language programming with open weights at $0.27/$1.12 per M tokens
Technical Specifications
Key Features
Capabilities
Other MiniMax Models
Explore more models from MiniMax
MiniMax-M2.5
MiniMax's flagship model matching frontier performance at 1/20th the cost with 80.2% SWE-bench Verified
MiniMax-M2.5-Lightning
Ultra-fast variant of M2.5 generating 100 tokens per second at $1/hour continuous operation
MiniMax-M2
MiniMax's general-purpose foundation model with strong tool use and deep search capabilities
