Progress over time
Pick up to eight benchmarks to see the best score on each, by the release date of the model that set it.
Model progress over time
In the past year, Humanity's Last Exam climbed the most, from 31.6% to 54.8%. GPT-6 Astra holds the best score now.
- Past 6 months
- Past year
- Past 2 years
- All time
- Each benchmark’s main source
- Independent runs only
- With labs’ own numbers
- Humanity's Last Exam (Center for AI Safety and Scale AI)
- SWE-bench Verified (Princeton NLP and OpenAI)
- SWE-Bench Pro (Scale AI)
- MMMU-Pro (MMMU team)
Sources: Official leaderboards, Epoch AI (CC BY 4.0) and labs’ model cards
| Benchmark | At the start | Now | Change | Holds the best score | New records | Scores from |
|---|---|---|---|---|---|---|
| Humanity's Last Exam | 31.6% | 54.8% | +23.2 pts | GPT-6 Astra | 4 | Official leaderboard |
| SWE-bench Verified | 73.5% | 83.5% | +9.9 pts | Claude Opus 4.7 | 3 | Epoch AI runs |
| SWE-Bench Pro | 43.6% | 61.5% | +17.9 pts | Muse Spark 1.1 | 4 | Official leaderboard |
| MMMU-Pro | 78.4% | 84.6% | +6.2 pts | GPT-5.6 Sol | 6 | Labs’ own numbers |
Filled dots are independent runs. Rings are numbers a lab reports about its own model. A line ends at the newest model its source has scored.