PerplexityapiPerplexity Fast Search
A cheaper, faster mode of the Perplexity Search API for developers whose agents make lots of web searches.
Details
- Turn it on by setting search_type to "fast" in a Search API request.
- Priced at $1 per 1,000 requests.
- Runs on Photon, a new retrieval and ranking service Perplexity wrote in Rust, which now handles ranking across Perplexity's search systems.
- Perplexity reports a 160 ms median response and 95% of results back in 230 ms or less.
- Across six agent benchmarks, Perplexity says it cut cost per task by 68% compared with the default search preset, at comparable quality.
- It is also now the default search in Hermes Agent for Nous Portal subscribers.
- The Python and TypeScript SDKs need a documented workaround to use it for now.
Best for
Agents and apps that run many web searches per task, where speed and cost matter more than full-page extraction.
When to use it
Search calls are slowing an agent down or driving up what each task costs.
Who benefits
Developers building research or automation agents on the Perplexity API.
Join the newsletter to get ahead with AI.
I share strategies, new modules, free guides, and tools to get you better at using AI in your work.
Leading a team? See team programs →