tulu-2-dpo-70B-ExPO: benchmark results
Provider: chujiezheng. Released 2024-04-26. Access: Open.
Aggregate Rating 1471 ± 25, rank #2823 of 10916 rated models, from 15 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - HellaSwag | 89.29 | Normalized accuracy (%) (10-shot) | 99 |
| Open LLM Leaderboard v1 - ARC Challenge | 72.7 | Normalized accuracy (%) (25-shot) | 93.1 |
| Open LLM Leaderboard v1 - MMLU | 69.61 | Accuracy (%) (5-shot) | 92.7 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 69.99 | MC2 (%) (0-shot) | 89.8 |
| Open LLM Leaderboard v1 - WinoGrande | 83.19 | Accuracy (%) (5-shot) | 87.5 |
| Open Chinese LLM - TruthfulQA MC | 58.77 | Accuracy (%) | 78.1 |
| Open Chinese LLM - ARC Challenge | 57.51 | Accuracy (%) | 75.3 |
| Open LLM Leaderboard v1 - GSM8K | 59.36 | Accuracy (%) (5-shot) | 71.2 |
| Open Chinese LLM - CMMLU | 58.18 | Accuracy (%) | 66.3 |
| Open Chinese LLM - HellaSwag | 61.49 | Accuracy (%) | 63.3 |
| AlpacaEval 2.0 | 25.72 | LC Win Rate (%) | 62.2 |
| Open Chinese LLM - C-Eval Semantic | 69.86 | Accuracy (%) | 51.2 |
Interactive version: theaggregate.ai/model?slug=tulu-2-dpo-70b-expo · How It Works · Data refreshed daily, snapshot 2026-11-08.