Artificial Analysis has made its MiniMax-M2.7 model available through six different inference providers, highlighting significant differences in speed and cost. SambaNovaAI leads in performance, achieving 435 tokens per second. Other providers like FireworksAI, Novita Labs, and Together AI are also part of this offering. AI
Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →
IMPACT Provides new options for inference, with SambaNovaAI setting a high bar for speed.
RANK_REASON This is a release of a model by a non-frontier lab, with performance metrics provided. [lever_c_demoted from research: ic=1 ai=1.0]