I benchmarked 4 web search APIs for AI agents on 30 live queries
Comments on “I benchmarked 4 web search APIs for AI agents on 30 live queries”
3 judges. Scored by all three judges. Only the day’s top stories get the claim check. How far stories are checked
A benchmark of four web search APIs (You.com, Tavily, Exa, Parallel) for AI agents, run on 30 live queries across GTM, insurance and finance use cases, with 3 repeats each.
Why it ranked
No. 154 of 155 with a score of 0.55. The panel rated it highest on new information (0.93) and substantiated (0.81), lowest on significance (0.46). The judges disagreed most on overclaim (spread 0.41), so the story is marked contested.
- AI judgeJudge 2claude-haiku-5-5
- Self-promotional benchmark of search APIs for agents: substantive small-sample evaluation with raw data, but heavy marketing links, n=30 queries, and vendor-run framing limit significance and breadth.
- AI judgeJudge 3gpt-5.6-luna
- A detailed, reproducible benchmark with raw artifacts, methodology, query-level evidence, and practical selection guidance. It is also a project launch with some promotional framing, but the empirical content is substantial and directly useful for AI-agent builders.
Each judge wrote its line after answering the rubric, with the site and the author still hidden. Every answer behind this is in the table under Score.
Score
0.55total under rubric v2.2Contestedjudges differ by 0.41 on overclaim
- Significance0.46
- Reader interest0.69
- Audience breadth0.50
- Practical relevance0.78
- Substantiated0.81
- Technical depth0.78
- New information0.93
- Counts against: Promotional0.31
- Counts against: Overclaim0.37
Each judge’s answer per question (0–1), their mean and spread, and what the question adds to the score (weight × mean).
| Question | Judge 1 | Judge 2 | Judge 3 | Mean | Spread | Weight | Contribution (weight times mean) |
|---|---|---|---|---|---|---|---|
| Significance | 0.55 | 0.25 | 0.58 | 0.46 | 0.33 | +0.24 | +0.111 |
| Reader interest | 0.76 | 0.50 | 0.82 | 0.69 | 0.32 | +0.26 | +0.180 |
| Audience breadth | 0.54 | 0.40 | 0.55 | 0.50 | 0.15 | +0.13 | +0.065 |
| Practical relevance | 0.96 | 0.60 | 0.78 | 0.78 | 0.36 | +0.22 | +0.172 |
| Substantiated | 0.99 | 0.60 | 0.84 | 0.81 | 0.39 | +0.05 | +0.041 |
| Technical depth | 0.99 | 0.60 | 0.73 | 0.78 | 0.40 | +0.05 | +0.039 |
| New information | 0.92 | 0.90 | 0.96 | 0.93 | 0.06 | +0.05 | +0.046 |
| Promotional | 0.10 | 0.45 | 0.38 | 0.31 | 0.35 | −0.19 | −0.059 |
| Overclaim | 0.59 | 0.35 | 0.18 | 0.37 | 0.41 , above the contested threshold | −0.11 | −0.041 |
| Sum of contributions | +0.553 | ||||||
| Claim-check penalty | 0.000 | ||||||
| Total, as published | 0.553 | ||||||
Spread is the highest minus the lowest judge answer. marks a spread above 0.40: the judges disagree, and the story is marked contested.
The badge for this ranking, for the author's page or anyone's: one line of HTML, and it links back here.
<a href="https://rubric.hrih.si/editions/2026-10-28/154/"><img src="https://rubric.hrih.si/badge/2026-10-28/154.svg" alt="Ranked #154 on Rubric, 28 Oct 2026" width="252" height="28"></a>Also covered by
No other coverage was clustered with this story.
Elsewhere
Elsewhere: HN 1 pts, not on front page
Recorded for context only, checked 28 Oct 2026, 06:47 UTC. It played no part in the ranking.