Baichuan-M2: benchmark results

Provider: Baichuan.

Aggregate Rating 1582 ± 44, rank #1224 of 10916 rated models, from 8 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
ClinicalMC (Chinese)45.5Mean of 11 task scores (%; triage accuracy, examination reca85.7
LogiMed-RoB - Verdict Reliability99.84Reliability (%)61.1
ClinicalMC (English)45.12Mean of 11 task scores (%; triage accuracy, examination reca55
LiveMedBench10.78Overall Score (%)48.6
LogiMed-RoB - Atomic Consistency23.68Fully consistent instances (%)33.3
LogiMed-RoB - Condition-Aware Ratio85.86Ratio (%)33.3
LogiMed-RoB - Domain Logical Fidelity56.91Fidelity (%)11.1
LogiMed-RoB - End-to-End Consistency0.8Fully consistent instances (%)11.1

Interactive version: theaggregate.ai/model?slug=baichuan-m2 · How It Works · Data refreshed daily, snapshot 2026-11-08.