Diffusion Language Models Explained — How Mercury Generates 1,000 Tokens Per Second
Mercury uses diffusion instead of autoregressive decoding to generate all tokens in parallel, hitting 1,000+ tokens/sec. We break down how it works.
Stay on top of AI and tech news, fresh research explained simply, and honest tool reviews.
Anthropic's Claude Mythos Leaked — A New AI Tier Nobody Was Supposed to See
Anthropic accidentally exposed Claude Mythos, a new Capybara-tier model above Opus with major cyber capabilities, via a …
The Four Color Theorem Now Runs in Near-Linear Time — First Improvement in 30 Years
A new paper by Kawarabayashi, Thorup, Mohar, and Thomassen gives an O(n log n) algorithm for 4-coloring planar graphs, …
Google's TurboQuant Compresses LLM Memory 6x With Zero Accuracy Loss — Here's How It Works
Google's TurboQuant algorithm compresses LLM KV cache memory by 6x with zero accuracy loss and no retraining needed. We …
China Bars Manus AI Founders From Leaving the Country — Meta's $2B Deal in Limbo
China has barred Manus AI founders from leaving the country as it reviews Meta's $2 billion acquisition. The AI cold war …
Claude Dispatch vs OpenClaw vs Google Mariner: Which Desktop AI Agent Is Worth Your Money?
Claude Dispatch, OpenClaw, and Google Mariner compared on pricing, setup, security, and daily use for desktop AI agent …
GitHub Copilot Will Train on Your Code by Default — How to Opt Out Before April 24
GitHub Copilot now uses Free, Pro, and Pro+ user interaction data for AI training by default. Here's what changed, what …
Donald Knuth's 'Claude's Cycles': When AI Solved a Problem the Father of CS Couldn't
Donald Knuth published Claude's Cycles after Claude Opus 4.6 solved his open graph theory problem in 31 tries. The full …
OpenAI Kills Sora: Disney's $1B Deal Dead, IPO Pressure Mounts
OpenAI shut down Sora just six months after launch, killing Disney's $1B partnership deal. Here's why it happened and …