Open LLM Leaderboard - IFEval: leaderboard
The "IFEval" subtask of Open LLM Leaderboard. A task from the Open LLM Leaderboard, the standard open-weight evaluation suite spanning knowledge, reasoning, math and truthfulness benchmarks.
Metric: Score. Source: huggingface.co. Saturation forecast: Estimated already saturated. 4576 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Llama 3.3 70B Instruct | 89.98 | #534 |
| 2 | Llama 3.1 70B Instruct | 86.69 | #561 |
| 3 | Qwen 2.5 72B Instruct | 86.38 | #454 |
| 4 | Rombos-LLM-V2.6-Qwen-14B | 84.32 | #516 |
| 5 | solar-pro-preview-instruct | 84.16 | #758 |
| 6 | Qwen2.5-14B-Instruct-1M | 84.14 | #578 |
| 7 | Mistral Large 2 (Nov) Instruct (2411) | 84.01 | #465 |
| 8 | EXAONE 3.5 32B Instruct | 83.92 | #837 |
| 9 | Llama-3.1-Tulu-3-70B | 83.79 | #482 |
| 10 | Qwen 2.5 32B Instruct | 83.46 | #498 |
| 11 | Qwen2.5-32B-Instruct-abliterated-v2 | 83.34 | #394 |
| 12 | Configurable-Llama-3.1-8B-Instruct | 83.12 | #815 |
| 13 | Llama-3.1-Tulu-3-70B-DPO | 82.82 | #729 |
| 14 | Llama-3.1-Tulu-3-8B | 82.67 | #849 |
| 15 | Awqward2.5-32B-Instruct | 82.55 | #371 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=open-llm-leaderboard-ifeval · How It Works · Data refreshed daily, snapshot 2026-10-20.