on-device AI

clipto ai video search 250m valuation a solid film reel disc

Clipto Uses AI to Search Terabytes of Video and Is Now Valued at $250M

Clipto, the three-year-old San Francisco startup whose on-device AI lets people search terabytes of their own video, audio, images and documents by describing what they want, has raised $15 million at a $250 million post-money valuation. This analysis covers the funding round and its investors, the roughly $15 million in annual recurring revenue and net-income profitability behind it, founder Henry Kang’s path from Carnegie Mellon robotics to a Tencent exit, the new Model Context Protocol connector that lets ChatGPT and Claude query a private media archive, Pro-plan pricing, how Clipto compares with Adobe, Apple, Google and TwelveLabs, and what the “memory layer” thesis means for businesses sitting on years of unused footage.

Read more
perplexity hybrid mode mac local models subtasks a solid piggy bank

Perplexity to Launch Hybrid Mode on Mac Using Local Models for Subtasks

TestingCatalog has surfaced Hybrid mode, an unreleased setting in Perplexity Computer’s Mac app that keeps orchestration in the cloud while delegating suitable subtasks to a model running locally, plus a Privacy Gate model that checks outbound data for personal information and asks the user before it is sent. This article covers the three downloadable local models (a 19GB Perplexity model, Qwen 32B at 17.4GB, and a 5.6GB Gemma-based model for 16GB Macs), which Mac mini configurations qualify, how Privacy Gate differs from June’s hybrid agentic inference design, the credit arithmetic the feature targets, how it compares with the Nvidia-only Portable Computer, and what a UK business should do about a feature that is still hidden and undated.

Read more
android app memory limits ai memory crunch a flat bar eight raised blocks

AI’s Memory Crunch Is Coming for Android Apps — Google’s New Memory Limits Explained

Google published two new Play app quality requirements on 27 August 2026, and the first of them makes memory usage a publishing condition from February 2027. Apps and games must stay under published per-tier thresholds for dynamic memory (anonymous RSS plus swap), bitmap memory and DEX code optimisation, measured at the 90th percentile over rolling 28-day windows. Google’s stated reason is unusual for a Play policy: “significant hardware supply constraints that are altering device memory availability” — the AI data centre boom draining consumer DRAM supply. This breakdown covers the published thresholds for apps and games across the 4 GB to 16 GB tiers, the Android 17 Memory Limiter that enforces the same idea through Linux cgroup v2, what tends to break first, the contradiction between Gemini Intelligence’s 12 GB requirement and shrinking budget-phone RAM, and the work to schedule before the deadline.

Read more
perplexity portable computer nvidia local ai agent a laptop open blank screen

Perplexity Partners With Nvidia to Launch Portable Computer, a Fully Local AI Agent With Zero Token Costs

Perplexity announced Portable Computer on 25 August 2026, built in close partnership with Nvidia: the full agentic Computer stack — model, inference engine, agent harness, tools, connectors and sandbox — running on your own hardware, where local work consumes no billing credits. It needs an Nvidia RTX GPU with at least 24GB of VRAM or a DGX Spark, runs on DGX OS or Ubuntu with Windows due in September 2026, and offers Qwen 3.8 27B, Perplexity’s post-trained PPLX 27B and, soon, Nvidia’s Nemotron 3.5 Lightning. This breakdown covers the hardware floor, the per-step cloud escalation flow, Perplexity’s own benchmark numbers, what zero token costs actually excludes, and who should pilot it now.

Read more
apple m5 ultra 80 core gpu ai 8k video a square chip package raised die

Apple’s M5 Ultra With an 80-Core GPU Will Power Through Your AI Models and 8K Video

  • August 25, 2026
  • AI News
  • Comment off

Apple announced the M5 Ultra on 25 August 2026 alongside a new Mac Studio: an up-to-80-core GPU with a Neural Accelerator in every core, up to 512GB of unified memory at 1.2TB/s, a 32-core Neural Engine and a media engine rated for 33 simultaneous streams of 8K ProRes 422. This breakdown separates the peak-theoretical claims from the measured workload figures, works out what 512GB actually buys you when you run large models locally, checks the 8K video claim against a real timeline, lists the prices and ship dates, and sets out the questions Apple did not answer.

Read more
small language models business a three ascending rounded pillars

Small Language Models: Complete 2026 Guide for Smart Teams

Small language models now handle most routine business AI work at a fraction of frontier-model cost. This guide maps the 2026 field — Phi-4-reasoning-vision, Gemma 4, Qwen 3.5 Small, Claude Haiku 4.5 and Ministral 3 — and shows exactly when a compact model beats a frontier LLM on cost, privacy and latency. It closes with a hardware plan and a 90-day deployment roadmap.

Read more
AI Smartphone Memory: How AI Is Reshaping India's Smartphone Market

AI Smartphone Memory: How AI Is Reshaping India’s Smartphone Market

Artificial intelligence is rapidly transforming the global smartphone industry. While consumers often focus on new AI-powered features such as intelligent photo editing, real-time translation, voice assistants, on-device image generation, and advanced productivity tools, manufacturers face an equally important challenge behind the scenes—memory capacity. Modern AI applications require significantly more RAM, faster storage, and higher memory […]

Read more
Apple Intelligence Approved for Launch in China with Alibaba’s Qwen AI

Apple Intelligence Approved for Launch in China with Alibaba’s Qwen AI

Apple Intelligence approved for launch in China marks a significant milestone for both Apple and the global artificial intelligence industry. After months of speculation surrounding Apple’s AI rollout in one of the world’s largest smartphone markets, reports indicate that Apple has received approval to introduce its AI features in China through a partnership with Alibaba […]

Read more

Tags

AIX AlmaLinux Alpine Linux authentication authorization Backup Recovery Boot Startup Cloud Integration Common Problems Cpu Memory Database Debian Debian 11 Debian 12 Bookworm Debian 13 Trixie DNS Firewall Fix Prevention FreeBSD FreeBSD 12 FreeBSD 13 FreeBSD 14 FreeBSD 15 Kernel Drivers Logging Monitoring Lvm Raid NetBSD Networking openSUSE Operations Oracle Linux Oracle Linux 8 Oracle Linux 9 Package Management Performance Proxy Security Hardening SLES Storage Disk Systemd Services Tls Ssl Troubleshooting ubuntu VPN windows server
CHAT