← All leaderboards
Web search
What this board tests: web search: a query in, ranked result links out. Every provider below ran the identical corpus on 2026-06-20. The composite blends measured quality, latency p50, cost per successful call and error rate. Method and corpora: how we measure.
| Rank | Provider | Composite | Recall@10 | Cost | Latency p50 | Field note |
|---|---|---|---|---|---|---|
| 1 |  Jina |
7.69/10 | 0.60 | $0.00033 / useful result | 1.4 s | First across all three query classes. |
| 2 |  You.com |
7.41/10 | 0.65 | $0.0077 / useful result | 1.4 s | |
| 3 |  Serper |
7.24/10 | 0.57 | $0.00053 / useful result | 1.5 s | |
| 4 |  Brave Search |
6.99/10 | 0.64 | $0.0078 / useful result | 1.3 s | |
| 5 |  Parallel |
6.76/10 | 0.64 | $0.0062 / useful result | 2.7 s | |
| 6 |  Exa |
6.7/10 | 0.61 | $0.014 / useful result | 1.6 s | |
| 7 |  Tavily |
6.66/10 | 0.58 | $0.014 / useful result | 2.1 s | |
| 8 |  SearchAPI.io |
6.44/10 | 0.57 | $0.007 / useful result | 2.5 s | |
| 9 |  SerpApi |
6.1/10 | 0.59 | $0.015 / useful result | 2.8 s | |
| 10 |  DataForSEO |
5.86/10 | 0.59 | $0.0034 / useful result | 2.9 s | |
| 11 |  Linkup |
5.47/10 | 0.39 | $0.013 / useful result | 3.2 s |
Low scores are published alongside high ones — a ranking you can only trust if nothing gets hidden. The gateway's fee sits on top-ups, not per call, so the order above earns us nothing either way.