OpenAI GPT-5.4 Launch - Humanity's Last Exam (with tools): leaderboard
The "Humanity's Last Exam (with tools)" subtask of OpenAI GPT-5.4 Launch. A benchmark result reported in the OpenAI GPT-5.4 launch announcement, as self-reported by the model developer.
Metric: Score (%). Source: openai.com. 4 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.4 Pro | 58.7 | #32 |
| 2 | GPT-5.4 | 52.1 | #79 |
| 3 | GPT-5.2 Pro | 50 | #61 |
| 4 | GPT-5.2 | 45.5 | #110 |
Interactive version: theaggregate.ai/benchmark?slug=openai-gpt-5-4-launch-humanity-s-last-exam-with-tools · How It Works · Data refreshed daily, snapshot 2026-10-20.