MINIMAX: The average daily token consumption of the M2 series text model in February 2026 has increased to over 6 times that of December 2025.

MINIMAX-WP announcement, February 2026, we released M2.5, which achieved global top-level performance in productivity scenarios such as programming, tool invocation, and office work. In terms of programming, M2.5 broke industry records in the SWE-BenchVerified test and improved efficiency by 37% compared to the previous generation M2.1. More importantly, M2.5 allows the operation of complex agents to become infinitely scalable in terms of economic cost - running continuously for an hour at a speed of 100 Tokens per second only costs 1 dollar, meaning that 10,000 dollars can allow 4 agents to work continuously for a year. The breakthrough in model capabilities has also driven a rapid increase in usage, quickly dominating OpenRouter after release. From M2 to M2.1 to M2.5, each generation of models has seen significant improvements in capabilities and usage. The average token consumption per day for the M2 series text models in February 2026 has increased to over six times that of December 2025, with token consumption from the programming package increasing by over 10 times.