Perplexity

A query — aware context compression module was deployed this month to the applications and API Platform.

Perplexity published Bumblebee, an open-source Go scanner for macOS and Linux that inspects developer endpoints for risky packages, editor and browser extensions, and AI agent configurations without executing installs

Perplexity’s Computer now connects to Snowflake, letting teams ask natural‑language questions against enterprise tables to surface metrics and act on insights without waiting on SQL, dashboards or analyst queues.

Perplexity has released Computer for Professional Finance, a finance — focused edition that combines licensed data connectors, built — in market tools, and 35 reusable workflows to produce source — traceable memos,

Perplexity Computer, an autonomous agent that writes and runs code, browses the web, and connects to external services, now runs on Perplexity’s infrastructure — which completed a 2026 SOC 2 Type II attestation —

A team put post‑trained Qwen3 235B into production on NVIDIA GB200 NVL72 racks built around Blackwell GPUs, adapting an in‑house inference engine to exploit tensor cores and high‑speed interconnects to maintain accuracy

Perplexity removed subscriber and waitlist limits for its Personal Computer macOS app, allowing any Mac user to run AI agents that access local files, native apps and the web.

Perplexity has integrated its Perplexity Computer architecture into a redesigned macOS application called Personal Computer, enabling teams of autonomous agents to run continuously on Macs and Mac mini while coordinating

Perplexity has introduced Finance Search in its Agent API, letting models fetch licensed datasets, real‑time market feeds and cited web sources in a single tool call so developers can return current, verifiable financial

Perplexity Research has added CuTeDSL kernels to its ROSE (Runtime — Optimized Serving Engine) to produce specialized GPU kernels and accelerate inference for large language and transformer models on NVIDIA Hopper

Computer, an AI agent that has logged over $2.8B in equivalent labor since launching a month ago, is now available inside Microsoft Teams.

Perplexity Research has released a public guide that codifies a folder‑based approach and distinct design principles for modular Agent Skills, explaining operational tradeoffs and offering examples from production use such as U.S.

Perplexity Hub has introduced 'Workflows for Computer,' a new suite of ready — to-run automations designed to simplify the application of AI for common business tasks.

Perplexity has significantly advanced its "Computer for Enterprise" platform with new integrations for Microsoft Teams and Excel, alongside an AI offering tailored for professional finance.

Perplexity Research described a two-stage post-training pipeline for search AI agents: SFT sets guardrails and format, while RL improves accuracy, efficiency, and quality of responses.

Perplexity has for the first time unveiled key technological approaches that ensure the high accuracy of its advanced AI models.