We finalized a new revision of MCP. Revision 2025-03-26 will bring Auth, Streamable HTTP, Audio modality and a few other goodies. We will be getting the SDKs up-to-date asap and will work towards a v 2.0 of the Python and Typescript SDK. But don't worry, everything is

people love MCP and we are excited to add support across our products. available today in the agents SDK and support for chatgpt desktop app + responses api coming soon!

We’re releasing PaperBench, a benchmark evaluating the ability of AI agents to replicate state-of-the-art AI research, as part of our Preparedness Framework. Agents must replicate top ICML 2024 papers, including understanding the paper, writing code, and executing experiments.

I am pleased to announce that the paper on our Dynamax Jax library for SSMs is now available at joss.theoj.org/papers/10.2110…. Code is at github.com/probml/dynamax/. Joint work with Scott Linderman Gerardo Duran-Martin ᴘᴇᴛᴇʀ ɢ. ᴄʜᴀɴɢ Aleyna Kara Giles HarperDonnelly Xinglong

Today is the start of a new era of natively multimodal AI innovation. Today, we’re introducing the first Llama 4 models: Llama 4 Scout and Llama 4 Maverick — our most advanced models yet and the best in their class for multimodality. Llama 4 Scout • 17B-active-parameter model

o3 and o4-mini are super good at coding, so we are releasing a new product, Codex CLI, to make them easier to use. this is a coding agent that runs on your computer. it is fully open source and available today; we expect it to rapidly improve.

I am pleased to announce that I have updated the online versions of my 2 textbooks (see probml.github.io/pml-book/): I fixed all issues listed on github, added some new references (esp on LLMs), and made a few other small tweaks.

x.com/i/article/1915…

Woke up feeling like a Premier League champion 🏆🔴

x.com/i/article/1916…

I am pleased to announce a new version of my RL tutorial. Major update to the LLM chapter (eg DPO, GRPO, thinking), minor updates to the MARL and MBRL chapters and various sections (eg offline RL, DPG, etc). Enjoy! arxiv.org/abs/2412.05265

Superb book - a must read from a world expert

x.com/i/article/1927…

Cursor IDE has been my friend for ~9 months now. But I find myself rarely using it any more. I'm addicted to Claude Code (CLI), and whilst it costs more, I get to work directly in the terminal - enjoying my time with vi again - and it's so much more performant. What do others

"everything that makes agents good is context engineering" excited to release dex's talk at AI Engineer, coiner of Context Engineering which has captured the zeitgeist of some of the most important problems in AI Engineering today!

Wrote up a few thoughts on Cursor's new $200/month Ultra plan and changes to their $20/month Pro plan simonwillison.net/2025/Jul/5/cur…

Rethinking multi-agent systems in the era of LLMs - 1 day workshop at Uni of Oxford - 16 Sept 2025 - Call for Expressions of Interest. (**Deadline 8 August**) sites.google.com/view/rethinkin…

Finally, a good modern book on causality for ML: causalai-book.net by Elias Bareinboim. This looks like a worthy successor to the ground breaking book by Judea Pearl which I read in grad school. (h/t Joshua Safyan for the ref).

Work life balance is a top priority. Yes, there was nearly a decade in my life when I worked 98 hours, but I did it out of need and aspiration, not because anyone forced me. During my Cambridge PhD I published more than anyone around me, but I never worked a single weekend. I

I've had preview access to GPT-5 for a couple of weeks, so I have a lot to say about it. Here's my first post, focusing just on core characteristics, pricing (it's VERY competitively priced) and interesting details from the GPT-5 system card simonwillison.net/2025/Aug/7/gpt…