SWE-Lancer Leaderboard
OpenAI's leaderboard for SWE-Lancer - 1,400+ real Upwork freelance software-engineering tasks totaling $1M in payouts, scored both technically and economically.
- Operator
- OpenAI
- Kind
- Aggregated
- Updates
- monthly
- Notable for
- The first major benchmark to map model performance directly to dollar earnings on real freelance work, complete with a public "SWE-Lancer Diamond" eval split.
- Tracks
- 1 evals · aggregated
Cite
Notes
Only stored in your browser.