Sarvam 105B

Sarvam 105B is a 105B+ parameter Mixture-of-Experts model with Multi-head Latent Attention, 128K context, Apache 2.0 open weights, and an OpenAI-compatible chat completions API. Sarvam positions it for complex reasoning, code generation, long-context document analysis, and agentic tool use. Official Sarvam materials report 98.6 on Math500, 88.3 on AIME 2025, 96.7 on AIME with tools, 49.5 on BrowseComp, and a 90% average win rate on Sarvam's Indian-language benchmark.