Skip to content

The Aggregate

Unified LLM rankings, daily

Models What's New Trends Benchmarks Skill Maps
How It Works Guesswork Metrics
The Aggregate
Loading data...

Open LLM Leaderboard - IFEval — leaderboard

Metric: Score. Source: huggingface.co. 4576 models tracked.

Top models

#ModelScore
1Llama 3.3 70B Instruct89.98
2Llama 3.1 70B Instruct86.69
3calme-2.1-qwen2.5-72B86.62
4Qwen 2.5 72B Instruct86.38
5calme-2.2-qwen2.5-72B84.77
6Rombos-LLM-V2.6-Qwen-14B84.32
7solar-pro-preview-instruct84.16
8Qwen2.5-14B-Instruct-1M84.14
9Mistral Large 2 (Nov) Instruct (2411)84.01
10EXAONE 3.5 32B Instruct83.92
11Qwen 2.5 32B Instruct83.46
12Qwen2.5-32B-Instruct-abliterated-v283.34
13zetasepic-abliteratedV2-Qwen2.5-32B-Inst-BaseMerge-TIES83.28
14Qwen2.5-14B-Instruct-abliterated-v283.28
15Configurable-Llama-3.1-8B-Instruct83.12

Interactive version: theaggregate.ai/benchmark?slug=open-llm-leaderboard-ifeval · How the rankings work · Data refreshed daily, snapshot 2026-07-22.

Built by Mikhail Doroshenko — AI researcher, co-author of Humanity’s Last Exam. Independent project; no affiliation with any AI lab.