Artificial Analysis Launches Search API Index: Parallel, Exa Lead
ArtificialAnlys · x · 2026-08-19
Artificial Analysis released a benchmark for search APIs, evaluating providers on quality, cost, and speed to help developers choose tools for AI agents. Initial coverage includes Parallel, Exa, Firecrawl, You.com, Tavily, and others.
Methodology:
- Uses the same model (GPT-5.6 Luna medium) within the open-source Stirrup agent harness.
- Only the search provider behind the tool changes.
- Scores are an average of DeepSearchQA, BrowseComp, and AA-Omniscience benchmarks.
Key Findings:
- Top Performers: Parallel, Exa, and Firecrawl lead with scores of 75, 74, and 73 respectively.
- Significant Lift: All search providers massively improved knowledge benchmarks (Model-only baseline: 33 vs. Search-assisted: 65-75).
- Cost Trade-offs: Higher quality search results reduce model inference token usage (Parallel Advanced saved >40% tokens vs Basic), offsetting higher search API costs.
Related event: Artificial Analysis Launches Search Index Benchmark for Search APIs(5 posts)→
More from coding & agent
- Apple code signing drags for hours; official advice: just wait a week — pjausovec · 2026-10-16
- Hermes: a 4B model with 8K context got worse with more test-time compute, pointing to contextual reasoning — mangahomanga · 2026-10-16
- Flux 3 Klein's Default Templates Are Broken Without Its Prompt Upscaler — StrictMove5839 · 2026-10-16
- 21-person team ships hundreds of PRs weekly with AI writing 97% of the code — rseroter · 2026-10-16
- ClosedMCP: Set Up MCP Servers Once, Share Them Across Your Whole Team — Ok_Sprinkles1301 · 2026-10-16
- Heisentools: let an agent rewrite a mock.py until it mirrors any real tool's API — kieranklaassen · 2026-10-16