Baichuan-2-7B-Base: benchmark results
Provider: Baichuan. Released 2023-09-06. Access: Open.
Unified ELO 1322 ± 1, rank #1321 of 1392 rated models, from 8 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Epoch AI - Lambada | 73.3 | Score | 36 |
| Big-Bench Hard | 41.6 | Average (%) | 27.1 |
| GSM8K | 24.6 | Accuracy (%) | 25.8 |
| MMLU | 54.16 | Accuracy (%) | 24.4 |
| HellaSwag | 68 | Accuracy (%) | 22.4 |
| InfiBench | 23.5 | Score (%) | 19 |
| BoolQ | 63.2 | Accuracy (%) | 14.3 |
| ARC Challenge (AI2) | 32.5 | Accuracy (%) | 6.6 |
Interactive version: theaggregate.ai/model?slug=baichuan-2-7b-base · How It Works · Data refreshed daily, snapshot 2026-09-05.