OpenAI
GPT-4o
One model for audio, images and text, answering in real time. In the API the same day.
- Announced
- Access
- Public
Benchmark scores
| Test | Score | Best on the tracker | How it was measured |
|---|---|---|---|
| Epoch Capabilities Indexmany tests combined into one score | 129 points | GPT-6.2 Astra 174 points | Independentlisted as "GPT-4o (Nov 2024)", CI 126.22 to 131.41Run 9 Nov 2026 |
| Humanity's Last Examexpert questions, no tools | 2.7% | GPT-6 Astra 55% | IndependentGPT-4o (November 2024), no tools, ±0.64Run 10 Apr 2025 |
| GPQA DiamondPhD-level science questionsPhD experts: 70%Saturated | 49% | Gemini 4 Argon 96% | Independentdefault settings (gpt-4o-2024-05-13)Run 27 Jan 2025 |
| OTIS Mock AIMEcompetition mathsSaturated | 6.3% | GPT-5.5 100% | Independentdefault settings (gpt-4o-2024-05-13)Run 25 Feb 2025 |
| METR time horizon (TH1.1 + TH2)task length an AI agent finishes half the time | 7.1 min | GPT-6 Astra 1.7 days | IndependentPooled TH1.1 + TH2 suites; 95% CI 3.8–13.5 min |
| METR time horizontask length an AI agent finishes half the timeNo longer updated | 7 min | Claude Mythos Preview 17 h | IndependentMETR Horizon v1.1 run 'gpt_4o_inspect'; 95% CI 4.0–12.9 min |
Also on these charts
Sources
- Hello GPT-4oOpenAI · Official · openai.com
- OpenAI API changelogOpenAI · Official · developers.openai.com
- Epoch AI Benchmarking Hub (model release dates and runs)Epoch AI · Dataset · 9 Nov 2026 · epoch.ai