Sources
Artificial Analysis launches a Search Index and the model-only baseline scores 33 against Perplexity's 80
The new leaderboard benchmarks 18 search API products across 8 providers as of 2026-08-27, scoring the equal-weighted mean of DeepSearchQA F1, BrowseComp exact-answer accuracy, and AA-Omniscience accuracy. Perplexity medium leads at 80 for $91.39 per 1,000 tasks at 28.6s each, ahead of Parallel advanced and Brave LLM-context both at 75, while running with no search at all scores 33 at $2.92 per 1,000 tasks. That spread is the first public number putting a price on retrieval quality for agent builders choosing a search backend.
↳ Follow the thread