Google A2UI 1.0 Replaces Agent Text With Screens
Google shipped A2UI 1.0 on November 11, 2026. And it attacks a problem anyone running an AI agent...
Claude Fable 5.2 Builds Playable Games From One Prompt
Claude Fable 5.2 launched on November 9, 2026. And it answers a question most people still treat as...
GitHub’s Git Storage Architecture Now Absorbs 5,000 Agents
GitHub’s Git storage architecture was rebuilt for a scenario that sounds made up: 5,000 agents pushing to one...
Non-Autoregressive Parallel Decoding: 10.5x Speed, One Big Catch
Non-autoregressive parallel decoding just cut decoding steps from 256 to 24 on MBPP, a 10.5× speedup reported by...
llama.cpp Can Now Compute Missed MoE Experts on the CPU
llama.cpp pull request 30966 gives the CPU an actual job in mixture-of-experts inference. When the GPU’s expert cache...
Agent-Native Primitives: What Actually Ships Your Code
Four agent-native primitives sit underneath Copilot, Claude. And Codex inside VS Code: the agent loop, the runtime that...
Persistent Memory For AI Coding Agents Just Went Mainstream
The changelog entry is almost boring. On October 28, 2026, GitHub announced that Copilot Memory is generally available...
Unslop: The Claude Skill That De-AIs Your UI
The unslop Claude skill audits five visual tells before it changes a single pixel: purple gradients, glows, emoji...
AI Agent Memory Promises 90% Token Savings. Here’s the Truth.
AI agent memory posted 91% lower p95 latency in Mem0’s own benchmark study, along with more than 90%...
Weight-Space Programming: The 0.016% Edit That Repaired Qwen
Qwen3.5-0.8B jumped from 7.9 to 48.1 percent on MATH-500 after a weight-space programming session edited 131,072 weights, roughly...