MiMo-V2.5-Pro: benchmark results

Xiaomi MiMo-V2.5 Pro open MoE model (42B active). Provider: Xiaomi. Released 2026-04-23. Access: Open.

Unified ELO 1671 ± 1, rank #95 of 1392 rated models, from 245 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
LLM Stats (CMMLU)90.2Score (%)100
LLM Stats (TAU3-Bench)72.9Score (%)100
AA IFBench79.86Accuracy (%)98
CLAW-Eval80.6Score (%)96
GACL - Tic-Tac-Toe81.04Normalized Score (0-100)95
Chatbot Arena (Text - Instruction Following)1471Arena Score93.5
Chatbot Arena (Text - English)1481Arena Score93.4
Chatbot Arena (Text - Coding)1520Arena Score93.1
Chatbot Arena (Text - Longer Query)1486Arena Score93.1
ClawProBench63.3Final Score (self-reported)92.9
LLM Stats (DROP)86.3Score (%)92.9
Chatbot Arena (Text - Hard Prompts)1496Arena Score92.4

Interactive version: theaggregate.ai/model?slug=mimo-v2-5-pro · How It Works · Data refreshed daily, snapshot 2026-09-05.