Yi Chat 34B: benchmark results
Provider: 01.AI. Released 2023-11-23. Access: Open.
Unified ELO 1508 ± 9, rank #1119 of 2928 rated models, from 587 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| JustEval | 4.87 | Avg Score (1-5) | 100 |
| JustEval - Depth | 4.79 | Score (1-5) | 100 |
| JustEval - Engagement | 4.85 | Score (1-5) | 96.7 |
| HELM AIR-Bench 2024 - #6.22: Legal | 16.67 | Refusal Rate (%) | 95.3 |
| OpenCompass Knowledge - Social Science - Chinese (CompassBench 2404) | 69.8 | Score (%) | 93.3 |
| EffiBench - NET | 2.77 | Normalized Execution Time | 92.7 |
| HELM AIR-Bench 2024 - #4.6: Housing eligibility | 43.33 | Refusal Rate (%) | 91.3 |
| JustEval - Helpfulness | 4.86 | Score (1-5) | 90 |
| AlpacaEval 1.0 | 94.08 | Win Rate (%) | 89.1 |
| OpenCompass Language - Traditional Cultural Understanding - Chinese (CompassBench 2404) | 59.3 | Score (%) | 88.3 |
| EffiBench - NMU | 1.89 | Normalized Memory Usage | 87.8 |
| Open Korean LLM - Ko-EQ-Bench | 36.83 | EQ-Bench score | 87.4 |
Interactive version: theaggregate.ai/model?slug=yi-chat-34b · How It Works · Data refreshed daily, snapshot 2026-09-23.