Yi-9B-200K: benchmark results
01.AI's 9B Yi base model with context extended to 200K tokens (March 2024), strong on code and math among similar-sized open models. Provider: 01.AI. Released 2024-03-16. Access: Open.
Aggregate Rating 1407 ± 20, rank #1204 of 1656 rated models, from 36 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - MMLU | 70.34 | Accuracy (%) (5-shot) | 93.8 |
| Open Portuguese LLM - OAB Exams | 50.57 | Accuracy (%) | 77.6 |
| Open Chinese LLM - C-Eval Semantic | 82.93 | Accuracy (%) | 73.7 |
| Open Chinese LLM - CMMLU | 62.82 | Accuracy (%) | 72.2 |
| Open LLM Leaderboard - GPQA | 31.54 | Score | 71.9 |
| Open LLM Leaderboard - MuSR | 42.94 | Score | 70 |
| Open Chinese LLM - HellaSwag | 62.01 | Accuracy (%) | 65.7 |
| Open Portuguese LLM - FaQuAD NLI | 71.62 | Macro F1 (%) | 64.1 |
| Open Portuguese LLM - ENEM | 65.64 | Accuracy (%) | 62.2 |
| Open LLM Leaderboard v1 - GSM8K | 47.61 | Accuracy (%) (5-shot) | 60.9 |
| HELMET (128K) | 33.9 | Average Score | 59.7 |
| Open Chinese LLM Leaderboard | 58.07 | Official Nonzero-Task Mean (%) | 58.9 |
Interactive version: theaggregate.ai/model?slug=yi-9b-200k · How It Works · Data refreshed daily, snapshot 2026-10-20.