MiniMax M2-her

MiniMax M2-her is a large language model with a 65K token context window, positioned as a lightweight tier within MiniMax's M2 generation.

Strengths

Offers fast inference and low latency suitable for real-time applications while maintaining competent performance across general language tasks.

Best for

Latency-sensitive deployments and scenarios where smaller model size and faster response time outweigh the need for extended context or maximum reasoning capacity.

Limitations

The 65K token context window is narrower than later models in the lineup such as M2.5 and M2.7, which support 204K tokens, limiting its ability to handle long documents or complex multi-turn conversations.