Baichuan4-Air: benchmark results
Provider: Baichuan.
Aggregate Rating 1519 ± 52, rank #2078 of 10916 rated models, from 16 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| K-12EduBench | 71.28 | Lambda_js Avg (imported) | 72.7 |
| ReLE - Education - Middle School Subjects | 34.7 | Accuracy (%) | 35.8 |
| ReLE - Education - High School Subjects | 42.4 | Accuracy (%) | 25.7 |
| ReLE - Reasoning - BBH | 72 | Accuracy (%) | 22.3 |
| ReLE - Education - Primary School Subjects | 42.2 | Accuracy (%) | 20.9 |
| ReLE - Education | 36.4 | Accuracy (%) | 19.9 |
| ReLE - Language and Instruction Following | 53.4 | Accuracy (%) | 13.5 |
| ReLE - Law - Lawyer Qualification Exam | 36.7 | Accuracy (%) | 11.6 |
| ReLE - Medicine and Mental Health | 55.3 | Accuracy (%) | 8.8 |
| ReLE - Finance | 57.4 | Accuracy (%) | 6.8 |
| ReLE - Medicine - Basic Medical Knowledge | 53.8 | Accuracy (%) | 6.8 |
| ReLE - Education - Gaokao | 26.4 | Accuracy (%) | 5.4 |
Interactive version: theaggregate.ai/model?slug=baichuan4-air · How It Works · Data refreshed daily, snapshot 2026-11-08.