Exa 3.0 ties Perplexity for first on Artificial Analysis Search Index
Overview
Artificial Analysis benchmarked Exa 3.0 with Dynamic Highlights and reports that its auto search mode scores 80 on the Artificial Analysis Search Index, tying Perplexity Search (medium) for first place.
Exa's lite mode scores 79. Artificial Analysis also reports that on DeepSearchQA, Exa's auto mode scores 85.6 F1 and lite 84.9 F1, ranking first and second and 4 to 5 points ahead of Perplexity.
Exa says Dynamic Highlights cuts searches per task from 9.4 to 7.6 and tokens per task by 47%, lowering cost per task 21% to $0.066. Exa prices auto at $7 per 1k searches and lite at $4.
Written by AI from the articles below · updated Oct 19, 1:57 PM ET
Check the sources:
Article timeline
The articles in this story. Times are ET.
Artificial Analysis@ArtificialAnlysOfficialExa 3.0 ties Perplexity for first on Artificial Analysis Search IndexAIExa releases Exa 3.0 with a Dynamic Highlights mode, where its auto search scores 80 on the Artificial Analysis Search Index, tying Perplexity Search (medium) for first place. Exa's lite mode scores 79, and on DeepSearchQA auto (85.6 F1) and lite (84.9 F1) rank first and second, 4 to 5 points ahead of Perplexity. Dynamic Highlights cuts auto's searches per task from 9.4 to 7.6 and tokens per task 47%, lowering cost per task 21% to $0.066; auto is priced at $7 per 1k searches and lite at $4.

Heat Trend
Not enough continuous observations to show a trend yet.