HREF — leaderboard
HREF evaluates instruction-following models with human response-guided automatic evaluation across 11 task categories.
Metric: Average HREF Score (%). Source: huggingface.co. Status: saturation imminent. 34 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Llama 3.1 70B Instruct | 48.98 |
| 2 | Mistral Large 2 (Jul) | 48.39 |
| 3 | Qwen 2.5 72B Instruct | 46.21 |
| 4 | Mistral-Small-Instruct-2409 | 42.87 |
| 5 | Qwen 1.5 110B Chat | 40.76 |
| 6 | Llama 3.1 8B Instruct | 38.57 |
| 7 | Yi 1.5 34B Chat | 35.1 |
| 8 | Qwen 2 72B Instruct | 33.71 |
| 9 | Llama-3.1-Tulu-3-8B | 33.54 |
| 10 | Phi-3-medium-4k-instruct | 30.91 |
| 11 | OLMo-2-1124-7B-Instruct | 28.49 |
| 12 | Llama 2 70B Chat (HF) | 23.9 |
| 13 | tulu-2-dpo-70B | 22.67 |
| 14 | Mistral 7B Instruct (v0.3) | 22.66 |
| 15 | Llama 2 13B Chat Base | 19.56 |
Interactive version: theaggregate.ai/benchmark?slug=href · How the rankings work · Data refreshed daily, snapshot 2026-07-22.