
Perplexity brings on-device Portable Computer agent to Windows RTX PCs
The local agent runs Qwen 3.8 27B on GeForce RTX and RTX PRO GPUs with 24GB+ VRAM, keeping sensitive files off the cloud.
Tag · 12 stories
Every story tagged Perplexity on AI Chat Daily.
Perplexity is the AI-native answer engine and maker of the Comet browser. Our Perplexity coverage follows product launches, enterprise moves, and the ongoing reshape of search as AI changes the category.

The local agent runs Qwen 3.8 27B on GeForce RTX and RTX PRO GPUs with 24GB+ VRAM, keeping sensitive files off the cloud.

The chipmaker is deepening ties with an AI search startup it already backs, cementing Nvidia's role as both supplier and shareholder across the sector.

Downloads collapsed 90% after the free offer closed, but monthly revenue hit $156,000 in July 2026 — more than four times January 2025.

Reddit was the most-cited domain by ChatGPT, Perplexity, Gemini, and Google AI Mode in May, drawing a wave of covert brand promotion.

The agentic tool works across local files, Microsoft 365, and the web, targeting enterprise workflows that live on Windows desktops.

The 88-core Arm chip aims to keep GPUs fed as agents run in loops; Perplexity ran a coding workflow 1.5x faster than x86.

CEO Aravind Srinivas tells CNBC the AI search startup will go public in 2028 even as Anthropic and OpenAI race to list first.

Anthropic, Perplexity, Runway, and Manus have all moved to serifs — a design choice critics are calling "tasteslop."

Snap returned to user growth with 483M DAU but lost its biggest generative AI revenue partner before it ever paid out.

The local AI agent app exits its Max-tier waitlist, runs on-device, and can be triggered remotely from an iPhone.

ASML's Christophe Fouquet says chips will be supply-constrained for two to five years as Google Cloud's backlog jumps from $250B to $460B in a quarter.

Ask.com closed on May 1, 2026, instead of being relaunched as an AI search wrapper. By any reasonable read of the comp set, that was the biggest unforced strategic error in consumer AI this year.
The tightly edited AI news email engineers, founders, and investors actually open. One email. Every weekday. Five minutes to finish.
The briefing read inside teams at