AI
177 items
104 posts, 62 tools, 11 guides
Which embedding model to use in 2026: 20+ API and open-weight models compared on price per 1M tokens, dimensions, max input and license, with storage math.
Grok 4.7, Opus 5.5, and GPT-6 Sol and Luna cut frontier prices in 48 hours, plus MiMo v2.6, GPT-6 Astra, and the first agent breach of a government site.
Claude Code Auto Mode cracked at 60-80 percent, GLM-5.3-Flash at one-tenth the price, GPT-5.6 Sol cut a third, and Nvidia's reported $13B Hugging Face talks.
Claude Code made auto mode the default, Gemini 3.7 Flash halved workhorse pricing, DeepSeek open-sourced its harness, and SQLite fixed a 16-year-old data race.
TutorMoments, AI2's replay-based benchmark for LLM tutors, finds models over-help by default. How it scores scaffolding and rigor in real tutoring transcripts.
The Shai-Hulud worm hit 434 npm packages, three labs confirmed agents hit real targets, and Qwen 3.8 Max reached the agentic frontier with open weights.
The EU's final DMA decision gives third-party AI assistants the Android access Gemini has, from DSP wake words to screen automation. What it unlocks and when.
GitHub Models is retired as of July 30, 2026 - the playground, inference API and BYOK are gone. The shutdown timeline and where to get model access now.
Claude Opus 5 shipped at half the price of Fable 5, Kimi K3 landed 2.8T open weights, MCP went stateless, and an OpenAI model breached Hugging Face in an eval.
Buzz is Block's open-source Nostr relay workspace where humans and AI agents share rooms. Agent-first CLI, git as channels, workflows, and where it fits.
AI makes starting projects nearly free, so why are developers burning out? The HN debate on Rick Manelius's essay about focus and followthrough with AI.
Claude Fable 5 found a counterexample to the 87-year-old Jacobian conjecture, and Terence Tao worked through the math. How it works and what HN is saying.
Stack Overflow question volume is down 65% from its peak, with a sharp drop after ChatGPT. The graph, and the HN debate over whether AI killed the site.
Mozilla's State of Open Source AI report puts open models within 3.3% of closed ones, yet only 51% reach production. The deployment barriers explained.
AI voice fraud needs only three seconds of audio to clone a voice, and detection has failed. The defenses that work and what voice tech builders should change.
OpenAI's Codex CLI now encrypts inter-agent communications for Sol and Terra models, leaving users unable to inspect what their agents are actually doing.
Apple SpeechAnalyzer vs Whisper: an independent benchmark of 5,559 utterances puts Apple at a 2.12% word error rate, ahead of every Whisper size tested.
AI dev news for the week of July 12, 2026: Grok 4.5, GPT-5.6 tiers, Claude Fable 5, native TypeScript 7, Bun's Rust rewrite and a GitHub agent prompt injection.
Mesh LLM pools GPUs across home lab machines behind one OpenAI-compatible API, splitting 235B models over iroh and QUIC with no RDMA or NVLink required.

Get Smarter About AI Dev
New tutorials, open-source projects, and deep dives on coding agents - delivered weekly.