arlen/benchOpen benchmarks for agentic consumers
SNAPSHOT web_search-2026-q3
11 JUN 2026 · BERKELEY, CA
Vendor Profile · Illustrative prototype

Serper

Serper on the arlen/bench leaderboards — agent web-search accuracy, extraction fidelity, freshness, cost per verified-correct answer, and agent-readiness, scored against golden truth.

Answer firstillustrative

Serper ranks #3 of 6 for agent web-search accuracy at 81.2% hit@5; #6 of 7 for extraction fidelity at 0.69, and is not yet agent-ready in live onboarding trials.

hit@5
81.2%
web search
cost / correct
$0.0041
web search
extraction fidelity
0.69
web extraction
agent-ready
No
onboarding harness
§ A

Web Search — Serper

snapshot web_search-2026-q3
Metric SerperLeaderboard bestDirection
hit@1 %66.0Exa 71.2higher is better
hit@5 %81.2Exa 87.6higher is better
fresh<30d %77.3Tavily 86.4higher is better
retrievability h10.6Tavily 7.8lower is better
cost/correct $$0.0041leadslower is better
p50 latency ms355Brave 301lower is better
§ B

Web Extraction — Serper

snapshot web_extraction-2026-q2
Metric SerperLeaderboard bestDirection
fidelity 0-10.69Firecrawl 0.91higher is better
JS gap Δ0.27Firecrawl 0.03lower is better
block rate %9.5Apify 3.3lower is better
cost/correct $$0.0027Jina $0.0009lower is better
schema validity %Firecrawl 96.2higher is better
§ C

Compare Serper

Serper vs Exa Serper vs Tavily Serper vs Brave Serper vs Firecrawl Serper vs SerpAPI

Illustrative prototype. No verified vendor run has been published yet; every figure here is a placeholder and must not be cited as a measured result. Numbers are replaced when a snapshot’s first full run lands.