Updated daily at 05:00 UTC
Benchmark
SimpleQA
Measures: The model answers short fact-seeking questions written adversarially against GPT-4. Each answer is graded correct, incorrect or not attempted, and the results are reported as % correct and an F-score.
- Maintainer
- OpenAI (original); a Kaggle-hosted leaderboard is still updated
- Introduced
- 2024
- Status
- Superseded
- Saturation
- Headroom
- Typical provenance
- Mixed
- In scoring set
- No
Leaderboard
Independent and third-party results only.
No independent or third-party results yet. Show all results