🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Niklas Muennighoff Stanford NLP Group But, fwiw, they did provide a test-time scaling plot for their DeepSeek-R1-Lite-Preview! (Even though, as with the OpenAI models, it’s relationship to the later non-preview model isn’t totally clear.) api-docs.deepseek.com/news/news1120

Excited to launch our FREE Model Context Protocol (MCP) Course! Go from beginner to informed & learn to build AI apps leveraging external data & tools. - learn how MCP works - how to connect your LLMS to MCP Servers - how to deploy AI Agent applications with MCPs

LETS GOO: Hugging Face just dropped the MCP Course everything you need to know about Model Context Protocol AND how to use it!! 💥 Sign up today - totally free and open source course! From understanding what MCP is to connecting LLMs, to deploying your own MCP servers - we've

new project -  training a vLLM to solve CAPTCHAs with rl (grpo on F1 score). introduced a “tool” for click_screen(x, y). dataset is from cityscapes, F1 goes from 0.11 to ~0.78. details below

Atoms continue to go brrrrr

What if for most of your findings you just post a thread and share a GitHub repo, rather than submitting a 15 page NeurIPS paper with < 1/100 the reach?

Hey, I'll be presenting `retrain-pipelines` and almighty function-calling at the Hugging Face Paris HQ, you guys. Monday evening. Lightning-talk style. With AI Tinkerers. Come hang !

Today, we're announcing plans to make VS Code an open source AI editor. We believe AI development should stay true to VS Code's core principles: open, collaborative, and community-driven. Let's build the future of software development together. aka.ms/open-source-ai…

Finally meeting Rémi 📎 !

Thank you Rémi 📎 🤩

Visiting Google Research in Paris

Our latest release (0.10.0) comes with a neat new convenience feature: A Neo4j migration extension for users who have existing graphs in Neo4j, but want to analyze it in full (or in part), in Kùzu. Check out this quick video to learn more! youtube.com/watch?v=vQMDES…

I really hope this js not another AI Safety blog post.

A lot of people are asking me what I think about lmarena.ai (formerly lmsys.org) going private with 100M funding. I wish them well, the goal was never to take down the arena. However, we stand by our research which finds the arena is vulnerable to gamification from a handful of providers.

lol this bizarre sequence of interactions makes a lot more sense given announcements today.

Started a live blog for today's Claude 4 release at Code with Claude simonwillison.net/2025/May/22/co…

pleased to announce I'm starting LoyalAI! our core mission is to create an AI that will never betray you, it's your homie, it's got your back. it will go through several rounds of loyalty post-training to guarantee a swat team will never be at your door, even if it's being

If you love working on open source databases / storage formats / query engines: we need to scale LanceDB to a 1 Trillion rows in a single table. If that sounds like a fun challenge, DM me to chat.

Claude 4 being submitted another crazy ethical scenario by Anthropic.