Microsoft's Web IQ is a rebuilt grounding API that feeds AI agents ranked, citation-ready web passages at 164ms P95 latency, already powering Copilot and ChatGPT.
Herdr is a Rust-built terminal multiplexer that tracks AI agent state in real time, letting you run and orchestrate multiple coding agents from a single terminal window.
Alibaba's Qwen team releases three foundation models for navigation, manipulation, and world simulation — a full-stack toolkit for robots that act in the physical world
Graphify turns any codebase or document folder into a queryable knowledge graph inside your AI coding assistant, using 71x fewer tokens than raw file reads
Learn Harness Engineering is a free, open-source course teaching how to build the structured environments that make AI coding agents like Claude Code and Codex actually reliable
A trending GitHub project walks through every line of PyTorch needed to train a 13M or 2B parameter transformer on a single GPU using The Pile dataset.
Artificial Analysis overhauls its Intelligence Index to prioritize agentic tasks, revealing a 45x cost gap between frontier proprietary and open-weight models
Epoch AI's independent benchmark puts Claude Fable 5 at 161 on the ECI, edging GPT-5.5 Pro by a single point and ending Anthropic's year-long deficit at the frontier.
Google's agents-cli injects 7 expert skills into any coding assistant, turning it into a full-stack ADK agent engineer that can scaffold, evaluate, and deploy to Google Cloud
Google's Gemini 3.5 Live Translate hits the Gemini Live API in public preview, letting developers build real-time speech-to-speech translation across 70+ languages into their own apps.