← All leaderboards
Browser actions
What this board tests: declarative in-page interactions (click, fill, scroll) in one call. Every provider below ran the identical corpus on 2026-09-14. The composite blends measured quality, latency p50, the track-specific cost metric and error rate. Method and corpora: how we measure.
| Rank | Provider | Composite | Task success | Cost | Latency p50 |
|---|---|---|---|---|---|
| 1 | 8.97/10 | 1.00 | $0.00098 / call | 5.4 s | |
| 2 | 8.62/10 | 1.00 | $0.005 / call | 6.7 s | |
| 3 | 8.41/10 | 1.00 | $0.002 / call | 11.9 s | |
| 4 | 7.47/10 | 0.80 | $0.0032 / call | 4.8 s | |
| 5 | 7.11/10 | 0.80 | $0.0009 / call | 23.3 s | |
| 6 | 5.7/10 | 0.60 | $0.001 / call | 41.6 s |
Low scores are published alongside high ones — a ranking you can only trust if nothing gets hidden. No per-call markup, no commission from any provider. Cost figures are priced by the 2026-09-14 run's own table; the price the gateway bills today is on each provider page.