Photon (Product): news timeline & CVEs

Photon held a funeral for mobile apps. Now it has $4.5M to help replace them with agents

Photon raised $4.5 million seed to let developers run agents over iMessage, WhatsApp, and other channels.

AI startup Photon, which helps developers run agents over messaging channels, raised $4.5 million in seed funding co-led by Gradient and A*, with participation from Vercel, HongShan, and others. Founders Daniel Tian and Ryan Zhu report more than 40,000 developer sign-ups, 10x revenue growth in four months, churn under 3%, and 5x messaging volume in the last month. The hosted service claims 99.95% uptime plus SOC 2 Type II and HIPAA compliance, while the open-source version still accounts for about 98% of use. Named users and integrations include Corgi, Boardy, Rho, Nous Research's Hermes agent, Tencent's QClaw, and LangChain.

TechCrunch · AIOct 1, 14:00 UTC · 19d agoAI industry2

Perplexity Introduces Photon: A Rust-Based Retrieval Engine That Cuts p99 Latency From 800 ms to 65 ms

Perplexity's Rust engine Photon cuts retrieval p99 latency from about 800 ms to 65 ms.

Perplexity introduced Photon, a Rust retrieval and ranking engine that now handles all production search traffic and replaces a forked open-source engine. The company says retrieval-and-ranking p99 latency fell from about 800 ms to about 65 ms, while Fast Search reports 160 ms at p50 and 230 ms at p95. Across 3,554 agent tasks, Fast scored 64.3% at an estimated $59.73 versus 64.0% at $187.60 for the default preset. Photon is available only as a hosted API at $1 per 1,000 fast-search requests and is not open source.

MarkTechPostSep 30, 08:26 UTC · 21d agoAI tools & infra2· 2 reads

[AINews] Opus 5.5 is good at explainer videos

Claude Opus 5.5 leads SimpleBench at 88.4% as GPT-6, Gemini 3.8 Flash, and MiMo-V2.6-Pro also ship.

Latent Space’s AINews recap says Claude Opus 5.5 shipped and leads SimpleBench at 88.4%, with Terminal-Bench-Science rising from 24% at low effort to 62% at xhigh. GPT-6 Astra and Opus 5.5 lead that benchmark by about 20 points over Fable 5.1, while Qwen3.8 Max scores 12%. Gemini 3.8 Flash posts 41 on the AA Intelligence Index at 291 tokens/s with 1M context, and MIT-licensed Xiaomi MiMo-V2.6-Pro scores 46. The issue also covers TypeSafe’s Jev decision model, LangChain agent tooling, and Perplexity’s Photon retrieval engine.

Latent SpaceSep 29, 02:44 UTC · 22d agoAI industry6· 2 reads

Photon

Timeline3 articles

Appears with

Photon held a funeral for mobile apps. Now it has $4.5M to help replace them with agents

Perplexity Introduces Photon: A Rust-Based Retrieval Engine That Cuts p99 Latency From 800 ms to 65 ms

[AINews] Opus 5.5 is good at explainer videos