OossaAI is evolving fast. We explain it simply.

Perplexity launches Photon engine, dropping p99 latency to 65 ms

Perplexity's new Rust‑based retrieval engine, Photon, reduces 99th‑percentile response time from 800 ms to 65 ms for its search API.

NoteOossa1 min read

Perplexity has rolled out Photon, a retrieval and ranking engine it built in Rust. The engine replaces a forked open‑source component and now handles all of Perplexity’s production search traffic. Photon also powers a new Fast Search mode in the Perplexity Search API, giving developers quicker answers. The company says the p99 latency – the time 99% of requests take – fell from about 800 ms to 65 ms.

Why it matters

Faster search responses make AI‑driven answers feel more immediate for users and can lower cloud costs for developers.

Was this article useful?
Share

Read next

Oossa · Newsletter

The week in AI, explained

Every Monday: the stories worth knowing, in plain language. Free, no spam.

Sources & references

#SourceOutletDateKey takeaway
1Perplexity Introduces Photon: A Rust-Based Retrieval Engine That Cuts p99 Latency From 800 ms to 65 ms ↗MarkTechPostSep 30, 2026Perplexity has released Photon, an in-house retrieval and ranking engine written in Rust.

1 sources

Last updated: ·Markdown·llms.txt