Yi 6B (Base): benchmark results
Provider: 01.AI. Released 2023-11-02. Access: Open.
Unified ELO 1334 ± 1, rank #1296 of 1392 rated models, from 84 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Chinese LLM - WinoGrande | 64.56 | Accuracy (%) | 67.4 |
| Open Chinese LLM - C-Eval Semantic | 74.77 | Accuracy (%) | 64.7 |
| Open Chinese LLM - CMMLU | 56.54 | Accuracy (%) | 61.7 |
| URIAL-Bench - STEM | 7.7 | Judge Score (0-10) | 61.1 |
| URIAL-Bench - Humanities | 8.78 | Judge Score (0-10) | 55.6 |
| EuroEval Portuguese NLU - MultiWikiQA PT | 69.73 | Reading comprehension Score (%) | 55.1 |
| URIAL-Bench - Roleplay | 6.95 | Judge Score (0-10) | 50 |
| MERA - ruHumanEval | 9.09 | pass@1 (%) | 45.4 |
| URIAL-Bench - Reasoning | 3.5 | Judge Score (0-10) | 44.4 |
| WinoGrande | 71.3 | Accuracy (%) | 42.5 |
| Open Chinese LLM - HellaSwag | 59.03 | Accuracy (%) | 42.1 |
| MERA - ruCodeEval | 5.49 | pass@1 (%) | 41.6 |
Interactive version: theaggregate.ai/model?slug=yi-6b-base · How It Works · Data refreshed daily, snapshot 2026-09-05.