agentIMAGE
Fast Search cuts cost per task by 68%, Perplexity says
Across six agent benchmarks, Fast Search cut cost per task by 68% versus the default preset at comparable quality, with a 160 ms median latency, the lowest of the search APIs compared.
Perplexity
@perplexity_aiArchived 2026.09.25Originally posted 2026.09.25
Perplexity @perplexity_aiOriginally posted 2026.09.25
@perplexitydevs Across the six agent benchmarks, it cut cost per task by 68% compared with the default preset, with comparable quality. At 160 ms, Fast Search had the lowest median latency of any search API we compared. https://t.co/UcQb9XiToy
x.com/perplexity_ai/status/2103184668313166085Archived 2026.09.25 11:56