URIAL-Bench - Overall — leaderboard
Metric: Judge Score (0-10). Source: huggingface.co. 19 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4 | 8.99 |
| 2 | GPT-3.5 Turbo | 7.94 |
| 3 | DBRX | 7.22 |
| 4 | Llama 2 70B Base | 7.11 |
| 5 | Mixtral 8x7B (v0.1) | 6.94 |
| 6 | Mistral-7B-v0.1 | 6.68 |
| 7 | Yi 34B (Base) | 6.68 |
| 8 | gemma-7B | 6 |
| 9 | Phi-2 | 5.85 |
| 10 | Llama 2 13B Base | 5.34 |
| 11 | Yi 6B (Base) | 4.97 |
| 12 | Llama 2 7B Base | 4.83 |
| 13 | gemma-2B | 3.97 |
| 14 | falcon-7B | 3.1 |
| 15 | mpt-7B | 1.49 |
Interactive version: theaggregate.ai/benchmark?slug=urial-bench-overall · How the rankings work · Data refreshed daily, snapshot 2026-07-22.