From the source
Microsoft AI introduced MAI-Thinking-1, a medium-sized reasoning model with 35B active parameters and ~1T total parameters using a sparse Mixture of Experts architecture.
The model achieves state-of-the-art performance on maths, knowledge, and coding benchmarks for its weight class, including 97.0% on AIME 2025 and 94.5% on AIME 2026, and is preferred over Claude Sonnet 4.6 in blind human side-by-side evaluations.
It is available in public preview through Microsoft Foundry with a 256k token context window, function calling, and enterprise-grade security and compliance.




