Microsoft unveils a sweeping AI overhaul at Build 2025, introducing a Copilot 'super app,' new AI reasoning models, a developer-optimized Windows 11 experience, and rewritten system components with support for next-gen silicon like Nvidia's RTX Spark.
MiniMax M3 launches as a groundbreaking open-weight AI model that outperforms GPT-5.5 and Gemini 2.5 Pro on key benchmarks, featuring a 1M token context window, native multimodal support, and autonomous capabilities powerful enough to independently reproduce a research paper over 12 hours, available now via desktop app and API starting …
NVIDIA launches its Model Card Generator toolkit, automating AI model documentation with a 91% completion rate and 76% accuracy in under a minute, using a three-stage pipeline to produce comprehensive model cards covering Bias, Explainability, Privacy, and Safety, with Oracle already deploying it across OCI AI infrastructure.
OmniParse launches as a fully local, open-source platform capable of converting over 20 unstructured data formats — including documents, images, audio, video, and web pages — into AI-ready markdown, deployable via Docker on a single T4 GPU and designed to power GenAI applications like RAG and fine-tuning.
A new standalone C++ engine called SAM3DBody-cpp now delivers real-time 3D full-body motion capture from a single camera with zero Python dependencies, supporting multi-person tracking, a 70-joint skeleton, and direct BVH export compatible with Blender.
Microsoft unveils a new family of AI coding models at its Build developer conference, aiming to reclaim market leadership in AI-assisted coding and reduce its dependence on OpenAI as competition from Anthropic, OpenAI's Codex, and Cursor intensifies.
Anthropic raises $65 billion in a Series H funding round at a near-trillion dollar valuation while simultaneously launching Claude Opus 4.8, a more honest and capable AI model that is four times less likely to miss its own coding flaws, with new agentic features and a teased next-gen model class …
Corporate AI spending faces a major reckoning as companies like Microsoft and Uber struggle to justify soaring costs and poor returns, with experts warning that unfocused deployment, high token costs, and poor use case selection are driving a correction away from AI overuse toward smarter, revenue-focused applications.
Microsoft is rolling out a faster, redesigned Microsoft 365 Copilot featuring a cleaner interface that loads twice as fast, smarter progressive disclosure tools, and deeper integration across desktop and mobile apps, letting users edit documents and ask questions directly from within paragraphs, spreadsheet cells, and slides.
Micrograd, a 150-line Python autograd engine with over 16,000 GitHub stars, brings PyTorch-like deep learning power to bare-bones code, making neural network backpropagation transparent and accessible at the most fundamental scalar level.