Two independent benchmarks, two different yardsticks — both put Parallel first for agent search

Asked:

“What is the best search for agents?”

Evidence as of 2026-11-17: 11 benchmark findings from two independent publishers — AIMultiple (24 configurations from six search APIs on 120 English questions, 2026-11-10) and the OpenBenchmarks family (45 constrained company-discovery questions, one corroborating publisher across several pages) — plus current vendor pages for the four parent search products. Scores are percentages, time in milliseconds or seconds as named, cost in USD.

The two verdicts, side by side — unlike metrics, so no shared axis

AIMultiple · accuracy benchmark · 2026-11-10

Winner: Parallel Advanced — 87.0% answer accuracy

Metric: answer accuracy (share of answers judged correct by ≥2 of 3 LLM judges) · 0–100% scale
24 configurations from six APIs (Brave, Exa, Firecrawl, Parallel, Perplexity, Tavily) on 120 English questions; fastest full configuration: Perplexity Fast Low at 180 ms p50.
OpenBenchmarks · multi-search agent test · 2026-09-22

Winner: Parallel basic — 46.5% F1

Metric: F1 (precision and recall together) on constrained company discovery · 0–100% scale, a far harder target than single-answer accuracy
45 company-discovery questions, several searches per task, same agent and tool budget for every API. Parallel basic: median task time 67.6 s, median task cost $1.12 (mostly model tokens), list price $0.005/search.

The 87.0% and 46.5% figures are not comparable: different tasks, different metrics, different scoring. What matters is the agreement — two publishers with different methodologies both rank Parallel first. “Best” still depends on whether your workload prioritises answer accuracy, raw latency, privacy, or search-plus-scraping.

What the verdicts cannot say alone

Latency specialists exist: AIMultiple measured Perplexity Fast Low as the fastest full configuration at 180 ms p50 — Parallel did not win on speed. aimultiple.com
Parallel basic is also the cheapest per call among OpenBenchmarks’ leaders at $0.005/search list price; task cost ($1.12 median) is dominated by model tokens, not search fees. openbenchmarks.com
When OpenBenchmarks lets the agent scrape as well as search, Exa deep leads instead (median task 95.8 s, $0.683, $0.012/search) — the winner changes with the tool loop. openbenchmarks.com
In OpenBenchmarks’ pure latency page, Exa instant leads (median task 49.8 s, $0.105) — ranking by speed and ranking by F1 crown different APIs. openbenchmarks.com

Recommendation: Parallel by default, specialists by workload

Vendor claims above are context, not independent benchmark evidence. Where latency or USD pricing is marked as not stated, the vendor page gives none.

Every benchmark finding in the evidence

Benchmark pagePublishedLeaderScore %Median task sTask cost $$/searchSource

Method: 11 benchmark-finding rows gathered as of 2026-11-17 from AIMultiple (published 2026-11-10) and the OpenBenchmarks benchmark family (2026-08 to 2026-09; one publisher, several pages of the same benchmark, treated as a single corroborating source — the 2026-09-22 “Best Web Search APIs for AI Agents” page carries the headline corroboration), plus one current vendor page per parent product for Parallel, Brave, Firecrawl and Tavily. Scores are percentages (answer accuracy or F1 depending on page); latency in milliseconds, task time in seconds, cost in USD. Two less relevant tool variants (Parallel Ingest API, Tavily CLI) appear only in vendor context, not as cards. Benchmark rankings shift with query mix, agent loop, fetching behaviour and evaluation metric — replay a representative private eval set before committing.

This report was generated automatically by Keenable SELECT at a user's request, from publicly available web sources linked herein. Keenable does not review, verify, or endorse its contents and makes no representation as to accuracy, completeness, or timeliness; AI-based extraction may contain errors. Nothing in this report is investment, legal, financial, or other professional advice. All trademarks and referenced content remain the property of their respective owners; no affiliation or endorsement is implied. To report an error, rights concern, or request removal: legal@keenable.ai.

Keenable SELECTAsk your own question
Made with Keenable SELECT