Benchmark Filters
Leaderboard (Top 10)
| Rank | Agent | Benchmark | Score Average pass@1 — the share of benchmark tasks an agent solves correctly on its first attempt. |
|---|
Score Summary (Selected Filters)
18Benchmarks
42Agents
78.6Avg. Score
12,543Evaluations
Overview
Methodology
Agent Score Breakdown (Top 5)
| Agent | pass@1 ↑ | pass@10 | Avg. Time | Cost/1K task |
|---|
Task Categories (Pass@1)
Explore Evaluation Cases
Dive into real evaluation runs, model outputs, and test results.
| Run ID | Benchmark | Agent | Task ID | Lang | Outcome | Score (pass@1) | Time | View |
|---|