Yi-1.5-6B-Chat: benchmark results
01.AI's Yi-1.5-6B-Chat, continuously pretrained from Yi on 500B additional high-quality tokens and fine-tuned on 3M diverse samples, improving coding, math and reasoning over the original Yi. Provider: 01.AI. Released 2024-05-13. Access: Open.
Aggregate Rating 1390 ± 11, rank #1259 of 1656 rated models, from 276 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - GSM8K | 67.1 | Accuracy (%) (5-shot) | 83 |
| OpenCompass Reasoning - Deductive - English (CompassBench 2405) | 20.9 | Score (%) | 82.9 |
| OpenCompass Language - Creation - Chinese (CompassBench 2407) | 46.6 | Score (%) | 79.4 |
| Open LLM Leaderboard - MuSR | 43.92 | Score | 77.4 |
| LatamBoard - Spanish PAWS | 61.5 | Score (%) | 72.7 |
| Open LLM Leaderboard v1 - MMLU | 64.24 | Accuracy (%) (5-shot) | 67.1 |
| OpenCompass Math - Primary School - English (CompassBench 2405) | 78.7 | Score (%) | 64.5 |
| Open LLM Leaderboard - MATH Level 5 | 16.24 | Score | 64.2 |
| LatamBoard - Spanish MGSM | 9.6 | Score (%) | 63.6 |
| OpenCompass Language - Dialogue - Chinese (CompassBench 2407) | 51.2 | Score (%) | 61.8 |
| Open LLM Leaderboard - IFEval | 51.45 | Score | 61.4 |
| Open PL LLM - PSC (generative, 5-shot) | 92.05 | Binary F1 (%) | 60.7 |
Interactive version: theaggregate.ai/model?slug=yi-1-5-6b-chat · How It Works · Data refreshed daily, snapshot 2026-10-20.