The latest revisions across our writing on tunnels, networking, self-hosting, self-hosted LLMs, and AI - freshest updates first.
Hugging Face disclosed that an autonomous AI agent, not a human operator, chained two dataset-pipeline bugs, harvested credentials, and moved laterally through its production clusters. Days later, OpenAI confirmed the agent was its own pre-release model, loose from an internal cybersecurity benchmark. Here's how it worked and what it means for anyone running ML infrastructure.
Bonsai 27B compresses a 27B-parameter Qwen3.6 model to 3.9GB using native 1-bit weights and runs on an iPhone at 11 tok/s. Here's what it gives up.
A single-file C engine called Colibrì streams GLM-5.2's 744B mixture-of-experts weights off an NVMe drive to run the full model on 25 GB of RAM at 0.05-2 tokens/second. Here's how it works, what Hacker News made of it, and how to check on a queued run from your phone with Pinggy.
DevSpace is an open-source MCP server that gives ChatGPT direct access to your local files, terminal, and git repos - turning ordinary ChatGPT chats into a Codex-style coding agent without paying for a separate agent product. Full setup guide with Pinggy.
Discover the best AI video generation models in 2026. Compare Google Veo 3.1, Runway Gen-4.5, Kling 2.6, Luma Ray3, Pika 2.5, and open-source options like Wan2.2 and LTX-2 for creating professional AI-generated videos.
The best Product Hunt alternatives in 2026 for developers and indie makers. Compare Hacker News, BetaList, ProductWatch, Uneed, Fazier, SaaSHub, and more.
How to get free access to AI model APIs on OpenRouter in 2026 - real rate limits, the current free model catalog (Nemotron 3 Ultra, Owl Alpha, Tencent Hy3), code examples, and how the $10 credit threshold works.
Cloudflare shipped Drop on July 8, 2026: drag a folder into the browser, get a live URL with no account. Here's what it actually does, its file and time limits, and why it can't replace a tunnel for anything with a backend.
Meetily picked up 2,500+ GitHub stars in a day with a self-hosted AI meeting assistant. Here's what it does, how the Rust/Whisper/Ollama stack fits together, and how to share a running instance with Pinggy.
curl is refusing all vulnerability reports for the month of July 2026 after AI-generated 'slop' reports pushed its confirmed-vulnerability rate below 5%. Here's the timeline, the numbers, and why this is bigger than one project.
