Trinity Mini (free) - API, Providers, Stats
Trinity Mini is a 26B-parameter (3B active) sparse mixture-of-experts language model featuring 128 experts with 8 active per token. Engineered for eff...
MiniMax-M2 is an agentic AI model optimized for complex tool use and reasoning across multiple steps.
It achieves state-of-the-art results in coding and agentic benchmarks through interleaved thinking and decision-making between tool interactions. The architecture utilizes a sparse Mixture-of-Experts approach with 230 billion total parameters and 10 billion active per token for efficient inference and high performance.