Yi-1.5-dolphin-9B: benchmark results
Cognitive Computations' Dolphin 2.9.1 Yi 1.5 9B, an uncensored fine-tune of 01.AI's Yi-1.5-9B trained on Dolphin's filtered instruction data with initial agentic and function-calling abilities. Provider: 01.AI. Released 2024-06-01.
Aggregate Rating 1416 ± 20, rank #1149 of 1640 rated models, from 30 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - MMLU | 68.12 | Accuracy (%) (5-shot) | 90.7 |
| Open Medical LLM - MMLU College Biology | 79.17 | Accuracy (%) | 79.7 |
| Open Medical LLM - PubMedQA | 75.2 | Accuracy (%) | 70.8 |
| Open Chinese LLM - WinoGrande | 64.8 | Accuracy (%) | 68.8 |
| Open Chinese LLM - CMMLU | 59.55 | Accuracy (%) | 68 |
| Open Chinese LLM - C-Eval Semantic | 77.93 | Accuracy (%) | 67 |
| Open Chinese LLM - HellaSwag | 61.7 | Accuracy (%) | 64.5 |
| Open Chinese LLM Leaderboard | 58.93 | Official Nonzero-Task Mean (%) | 64.2 |
| Open Medical LLM - MMLU College Medicine | 64.74 | Accuracy (%) | 64.1 |
| Open Portuguese LLM - OAB Exams | 45.51 | Accuracy (%) | 61.1 |
| Open Portuguese LLM - BLUEX | 54.38 | Accuracy (%) | 59.8 |
| Open LLM Leaderboard v1 - GSM8K | 44.58 | Accuracy (%) (5-shot) | 58.6 |
Interactive version: theaggregate.ai/model?slug=yi-1-5-dolphin-9b · How It Works · Data refreshed daily, snapshot 2026-11-07.