Yi-6B-200K: benchmark results
01.AI's Yi-6B-200K, a long-context variant of the bilingual Yi-6B base model extending support to a 200,000-token context, trained from scratch on a 3-trillion-token corpus. Provider: 01.AI. Released 2023-11-05. Access: Open.
Aggregate Rating 1297 ± 39, rank #1488 of 1656 rated models, from 35 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard - MuSR | 45.87 | Score | 87.2 |
| Open Portuguese LLM - OAB Exams | 43.28 | Accuracy (%) | 54.9 |
| HELMET (128K) | 27.8 | Average Score | 46.8 |
| MERA - ruCodeEval | 5.73 | pass@1 (%) | 43.3 |
| MERA - ruHumanEval | 8.6 | pass@1 (%) | 43.1 |
| Open Portuguese LLM - BLUEX | 46.73 | Accuracy (%) | 38.1 |
| Open Portuguese LLM - ENEM | 54.23 | Accuracy (%) | 37.3 |
| Open LLM Leaderboard - GPQA | 28.19 | Score | 36.8 |
| Open LLM Leaderboard - MMLU-Pro | 28.44 | Score | 32.1 |
| Open LLM Leaderboard - BBH | 42.89 | Score | 29.3 |
| MERA - MathLogicQA | 35.87 | Accuracy (%) | 26.2 |
| MERA - SimpleAr | 90.2 | EM (%) | 21.9 |
Interactive version: theaggregate.ai/model?slug=yi-6b-200k · How It Works · Data refreshed daily, snapshot 2026-10-20.