Every story across every category, newest first. Each card links to the original publisher; daily-brief posts open as editorial pages.
QWEN CHAT GITHUB HUGGING FACE MODELSCOPE DISCORD We are excited to introduce Qwen-Image-Edit, the image editing version of Qwen-Image. Built upon our 20B Qwen-Image model, Qwen-Image-Edit…
Liquid AI, the MIT-born leader in efficient foundation models, today announced native support for AMD’s latest Ryzen™ and Ryzen AI™ processors within the Liquid Edge AI…
Anthropic gives Claude Opus 4 and 4.1 the ability to end abusive conversations — "model welfare" feature, first major lab to ship such a safeguard. ChatGPT…
A quiet Saturday dominated by the rolling aftermath of OpenAI's GPT-5 rollout. Weekend coverage of Altman's SF reporters dinner spills out — admissions OpenAI "totally screwed…
Sam Altman tells reporters at SF dinner OpenAI will spend "trillions of dollars" on data centers in the "not very distant future" — admits OpenAI "totally…
Financial teams are unlocking a new era of compliance, efficiency, and customer trust—powered by AI agents.
Insights and best practices for using Elo scoring methods for evaluation leaderboards.
Cohere raises $500M at $6.8B valuation led by Radical Ventures and Inovia with NVIDIA, AMD Ventures, Salesforce Ventures participating. Joëlle Pineau (ex-Meta FAIR head) joins as…
Fresh funding enables Cohere to accelerate its global expansion and build the next generation of secure enterprise and sovereign AI solutions.
DeepWiki, Vibe and Replace, Dev Containers, and more!
WRI and the Bezos Earth Fund used DINOv3 to develop an algorithm to accurately count individual trees from drone and satellite imagery.
Using DINOv2, the team at NASA's Jet Propulsion Laboratory built a convenient robot operating system interface for robotic tasks.
DINOv3 scales self-supervised learning for images to create universal vision backbones that achieve absolute state-of-the-art performance across diverse domains, including web and satellite imagery.
Bloomberg's Mark Gurman publishes Apple AI-comeback scoop — 2027 tabletop robot (7-inch display on a motorized arm), smart-home hub for 2026, home-security cameras, lifelike Siri with…
Anthropic offers Claude to all three branches of US government for $1/year via a GSA OneGov deal — matches OpenAI's federal land-grab from a week earlier.…
Announcing Deep Ignorance: Filtering Pretraining Data Builds Tamper-Resistant Safeguards into Open-Weight LLMs
Today, we release LFM2-VL, our first series of vision-language foundation models. These multimodal models are designed for low-latency and device-aware deployment. LFM2-VL extends the LFM2 family…
OpenAI quietly restores GPT-4o for ChatGPT Plus via a "show legacy models" setting — Altman concedes the company underestimated how much users valued GPT-4o's tone and…
Sam Altman publishes a long X thread on users' emotional attachment to specific AI models — justifying OpenAI's decision to bring GPT-4o back for Plus subscribers.…
Use our MCP to discover, compare, and run models from apps like Claude, Cursor, and VS Code.
Altman spends the weekend on social media trying to quell the GPT-5 uproar. Model router blamed for making GPT-5 "seem way dumber" — autoswitcher "broke and…
And How They Stack Up Against Qwen3
Throughout 2025, we have been quietly entering Claude in cybersecurity competitions designed primarily for humans. In many of these competitions Claude did pretty well, often placing…
GPT-5 day-two damage control. Sam Altman runs a Reddit AMA, doubles GPT-5 rate limits for Plus (160 msg/3h), restores GPT-4o for Plus subscribers, apologizes for the…