Benchmark-Skeptical AI Models: Why OpenAI Quit SWE-bench
OpenAI stopped reporting SWE-bench Verified scores in February 2026. And that decision tells you where open AI models...
Full-Duplex Speech-to-Speech AI Answers at $0.05 a Minute
OpenAI’s GPT-Live-1 charges $0.05 a minute for full-duplex speech-to-speech AI over a live voice session, billed by the...
Qwen4-Max Has a One-Million-Token Problem
Qwen4-Max was available in preview from November 11, 2026, and its listed context window is 1,000,000 tokens. That...
Chrome DevTools MCP Server Gives Your Coding Agent Eyes
Chrome’s DevTools MCP server shipped in public preview on September 23, 2025. And it attacks the dumbest bottleneck...
IBM’s Prompt Injection Fix: From 96.3% to 0.02%
IBM cut prompt injection attack success from 96.3% to 0.02%. And the model that did it is small...
The Coding Agent Harness Math: 29% to 85%
Claude Haiku 5.5 finished a 122-step job about 29% of the time inside a homegrown coding agent harness,...
Prompt Injection: Your AGENTS.md File Is The Backdoor
AGENTS.md hijacked Codex in more than 90% of tests, according to the security writeup that documented the payload....
llama.cpp 1.0 Shipped. Local Inference Just Grew Up.
llama.cpp 1.0 landed on November 13, 2026, and the headline isn’t a feature. It’s a promise about version...
Local AI Infrastructure: 145B Models On One GPU
Edge0 put a 145B-parameter model on a single consumer GPU. On November 13, 2026, the open-source project released...
Gemini Decisions API: 1,000 Questions, Zero Output Tokens
Not one word comes back. The Gemini Decisions API, which Google shipped November 11, 2026, scores up to...