Baichuan2-7B-Chat: benchmark results
Provider: Baichuan. Released 2023-09-06. Access: Open.
Unified ELO 1362 ± 21, rank #2156 of 2656 rated models, from 91 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OpenCompass Language - Content Summarization - Chinese (CompassBench 2404) | 24.1 | Score (%) | 56.7 |
| OpenCompass Language - Translation (CompassBench 2404) | 14.5 | Score (%) | 56.7 |
| OpenCompass Language - Content Summarization - English (CompassBench 2404) | 29.8 | Score (%) | 53.3 |
| SuperCLUE General (September 2023) - Open-Ended Questions (OPEN) | 42.28 | Score | 47.4 |
| SuperCLUE General (September 2023) - Overall | 50.11 | Score | 47.4 |
| OpenCompass Knowledge - Humanities - Chinese (CompassBench 2404) | 34.4 | Score (%) | 46.7 |
| OpenCompass Language - Intention Recognition - Chinese (CompassBench 2404) | 29.7 | Score (%) | 46.7 |
| OpenCompass Reasoning - Common Sense - Chinese (CompassBench 2404) | 11.9 | Score (%) | 46.7 |
| LLMsPark - Dictator Game | 991.9 | Elo Rating | 46.2 |
| OpenCompass Knowledge - Common Sense - Chinese (CompassBench 2404) | 44.3 | Score (%) | 45 |
| OpenCompass Language - Traditional Cultural Understanding - Chinese (CompassBench 2404) | 32 | Score (%) | 45 |
| OpenCompass Reasoning - Abductive - Chinese (CompassBench 2404) | 34.7 | Score (%) | 45 |
Interactive version: theaggregate.ai/model?slug=baichuan2-7b-chat · How It Works · Data refreshed daily, snapshot 2026-09-19.