dolphin-2.9.1-yi-1.5-34B: benchmark results
Eric Hartford's Cognitive Computations uncensored Dolphin fine-tune of Yi 1.5 34B with coding and function-calling skills. Provider: Cognitive Computations. Released 2024-05-18. Access: Open.
Aggregate Rating 1508 ± 22, rank #760 of 1656 rated models, from 25 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - MMLU | 77.52 | Accuracy (%) (5-shot) | 98.8 |
| Open LLM Leaderboard v1 - GSM8K | 73.01 | Accuracy (%) (5-shot) | 98.4 |
| Open Chinese LLM - GSM8K | 67.93 | Accuracy (%) | 92.3 |
| Open Chinese LLM - C-Eval Semantic | 89.13 | Accuracy (%) | 90.2 |
| Open Portuguese LLM - FaQuAD NLI | 79.81 | Macro F1 (%) | 88.4 |
| Open LLM Leaderboard - MuSR | 45.98 | Score | 87.7 |
| Open Chinese LLM - WinoGrande | 68.59 | Accuracy (%) | 87.4 |
| Open Chinese LLM Leaderboard | 67.46 | Official Nonzero-Task Mean (%) | 87 |
| Open Chinese LLM - HellaSwag | 67.88 | Accuracy (%) | 85.5 |
| Open LLM Leaderboard - GPQA | 34.31 | Score | 85.4 |
| Open LLM Leaderboard v1 - WinoGrande | 82.56 | Accuracy (%) (5-shot) | 84.6 |
| Open LLM Leaderboard - MMLU-Pro | 45.19 | Score | 84.1 |
Interactive version: theaggregate.ai/model?slug=dolphin-2-9-1-yi-1-5-34b · How It Works · Data refreshed daily, snapshot 2026-10-20.