JustVugg/colibri is a pure-C, zero-dependency inference engine by Vincenzo Fornaro that streams routed MoE experts from NVMe, running models like GLM-5.2 and Kimi K3 on consumer hardware. 51,687 stars as of 17 November 2026.
Read ArticleExplore our latest tutorials, tech updates, and career advice written by industry experts to help you level up your skills.
JustVugg/colibri is a pure-C, zero-dependency inference engine by Vincenzo Fornaro that streams routed MoE experts from NVMe, running models like GLM-5.2 and Kimi K3 on consumer hardware. 51,687 stars as of 17 November 2026.
Read ArticleSeer (arXiv:2611.26423) is an inference-time framework for small deep-research agents. The paper reports GAIA accuracy rising from 13.6% to 41.7% on Qwen3.5-4B and S32 reaching 79.6% on GAIA.
Read ArticleBaidu released ERNIE 5.5 Flash Base on 5 November 2026 — a sparse mixture-of-experts base model with 280 billion total parameters and 14 billion active per token, with weights downloadable from Hugging Face.
Read ArticleScrapling (D4Vinci/Scrapling) is a Python web-scraping framework and one of the most-starred scraping tools on GitHub, at roughly 93,000–100,000 stars as of 21 November 2026. Its rise reflects how much scraping work now feeds AI data pipelines.
Read Articleclaude-mem (thedotmack/claude-mem) gives coding agents persistent memory across sessions, so the agent remembers earlier decisions and context. About 115,000 stars as of 21 November 2026; one of the fastest-growing agent-tooling projects of the month.
Read ArticleNVIDIA's auto-ontology project builds an ontology/semantic layer over data so AI agents can reason about what data means, not just query it. About 10,500 stars as of 21 November 2026.
Read Articlevercel-labs/agent-desktop is a CLI/C library that lets AI agents operate desktop applications through the operating system's accessibility tree — reading the UI and driving clicks and keystrokes the way a human would. The desktop counterpart to the browse
Read ArticleMIT CSAIL's "Sparks of Weight-Space Programming" (2 Nov 2026): GPT-6 Astra edited 0.016% of Qwen3.5-0.8B's weights — no training loop — and MATH-500 accuracy rose from 7.9% to 48.1%. A repair, not new knowledge.
Read ArticleGoogle DeepMind announced AlphaProof Code on 10 Nov 2026: five researchers with Gemini 4 Carbon agent teams produced ~6.5M lines of machine-checked Lean proof for Signal's libsignal library — with clear limits the researchers state themselves.
Read ArticleAsk about courses, website & more
Request human callback:
Please log in to download. You will stay on this page after logging in!
New here? Register now →