OpenAI internal data-science tasks - reported score
Science
Internal data-science evaluation reported by OpenAI. Task roster and metric are undisclosed; diagnostic.
Top models (higher is better)
| Model | Score |
|---|---|
| GPT-6 Astra | 40.9 |
Benchmarks
218 sources · 1,060 models × 2,130 benchmarks · 109,661 scoresScience
Internal data-science evaluation reported by OpenAI. Task roster and metric are undisclosed; diagnostic.
| Model | Score |
|---|---|
| GPT-6 Astra | 40.9 |