Independent AI researcher, creator of datasette.io and llm.datasette.io, building open source tools for data journalism, writing about a lot of stuff at https://simonwillison.net/https://bsky.app/profile/simonwillison.nethttps://bsky.app/profile/simonwillison.net/post/3msc7zqolvk2bBig new release of my LLM CLI tool and Python library for talking to hundreds of different LLMs - reasoning traces, OpenAI Responses support, server-side tools, smarter logging and a whole lot more https://simonwillison.net/2026/Aug/4/new-release-of-llm/05 Aug 2026 00:03 +0000at://did:plc:kft6lu4trxowqmter2b6vg6z/app.bsky.feed.post/3msc7zqolvk2bhttps://bsky.app/profile/simonwillison.net/post/3msbpz3kmak27The MiniMax-H3 video generation model is a lot of fun - here's what I generated on my M5 Pro Mac for the prompt "a rainbow colored skunk leaps over a mossy log in a supermarket" (~115GB model download, took around 45 minutes to generate) - notes on how I ran it here https://simonwillison.net/2026/Aug/4/minimax-h3-mlx/04 Aug 2026 19:16 +0000at://did:plc:kft6lu4trxowqmter2b6vg6z/app.bsky.feed.post/3msbpz3kmak27https://bsky.app/profile/simonwillison.net/post/3ms7ccvfvbk2rQuestion for dog companions: are you aware that "nose work" is a sport you can participate in with your dog? We've been doing it with Cleo and it is ABSURDLY fun. Dogs have to find 3 different official hidden scents (clove, anise, birch) - you have to accompany them and call it when they find them03 Aug 2026 20:06 +0000at://did:plc:kft6lu4trxowqmter2b6vg6z/app.bsky.feed.post/3ms7ccvfvbk2rhttps://bsky.app/profile/simonwillison.net/post/3ms42wydw4k2hHere's my attempt at summarizing the various "open letters" about AI development that have been doing the rounds over the past few weeks https://simonwillison.net/2026/Aug/2/open-letters/02 Aug 2026 13:16 +0000at://did:plc:kft6lu4trxowqmter2b6vg6z/app.bsky.feed.post/3ms42wydw4k2hhttps://bsky.app/profile/simonwillison.net/post/3mry67nj6jk25Got a disappointing pelican from DeepSeek-V4-Flash-0731 at default reasoning mode - on the left - but then I bumped reasoning up to high (via OpenRouter) and got the much better one on the right https://simonwillison.net/2026/Jul/31/deepseek-v4-flash-0731/01 Aug 2026 00:04 +0000at://did:plc:kft6lu4trxowqmter2b6vg6z/app.bsky.feed.post/3mry67nj6jk25https://bsky.app/profile/simonwillison.net/post/3mry3ilyu7s2fThe new stateless MCP specification has rekindled my interest in MCP, and inspired some new projects, including mcp-explorer and datasette-mcp https://simonwillison.net/2026/Jul/31/stateless-mcp/31 Jul 2026 23:15 +0000at://did:plc:kft6lu4trxowqmter2b6vg6z/app.bsky.feed.post/3mry3ilyu7s2fhttps://bsky.app/profile/simonwillison.net/post/3mrxxwdfhxk2oI'm on Oxide and Friends podcast this week! We talked about accidental cyberattacks, Kimi K3, Golden Gate Claude, Zizians, Alameda wild turkey attacks, Soviet Marburg virus research, the Lead-crime hypothesis, and a bunch of other worthy digressions https://oxide-and-friends.transistor.fm/episodes/the-open-weight-revolution-with-simon-willison31 Jul 2026 22:11 +0000at://did:plc:kft6lu4trxowqmter2b6vg6z/app.bsky.feed.post/3mrxxwdfhxk2ohttps://bsky.app/profile/simonwillison.net/post/3mrxwydoc5k25I've been working with Prime Radiant building a new tool for running small eval suites against models, harnesses, and prompts - it's called "smevals", you can try it with "uvx smevals docs", and I wrote about it here: https://primeradiant.com/blog/2026/smevals.html31 Jul 2026 21:54 +0000at://did:plc:kft6lu4trxowqmter2b6vg6z/app.bsky.feed.post/3mrxwydoc5k25https://bsky.app/profile/simonwillison.net/post/3mrx6kneuhc2iYour infrequent reminder that if you live near the San Francisco Bay Area you should go to Moss Landing (south of Santa Cruz) and rent kayaks there, because the estuary is full of sea otters and you will encounter dozens of them31 Jul 2026 14:37 +0000at://did:plc:kft6lu4trxowqmter2b6vg6z/app.bsky.feed.post/3mrx6kneuhc2ihttps://bsky.app/profile/simonwillison.net/post/3mrr3z77wj224A new TIL on adding custom MCP servers to both the ChatGPT and Claude regular chat interfaces - it's a little less obvious than I had hoped, but I got there in the end https://til.simonwillison.net/llms/mcp-in-claude-and-chatgpt29 Jul 2026 04:36 +0000at://did:plc:kft6lu4trxowqmter2b6vg6z/app.bsky.feed.post/3mrr3z77wj224https://bsky.app/profile/simonwillison.net/post/3mrqeabti3s2yHugging Face just published a highly detailed technical account of OpenAI's accidental cyberattack on their systems - it's wild how sophisticated this was: https://huggingface.co/blog/agent-intrusion-technical-timeline Wrote up some of my own notes here: https://simonwillison.net/2026/Jul/28/anatomy-of-a-frontier-lab-agent-intrusion/28 Jul 2026 21:30 +0000at://did:plc:kft6lu4trxowqmter2b6vg6z/app.bsky.feed.post/3mrqeabti3s2yhttps://bsky.app/profile/simonwillison.net/post/3mriy5j5l322jRuff 0.16.0 - Astral's fast Python linter - came out a few days ago and increased the number of default-enabled rules from 59 to 413, which highlighted all sorts of problems across my projects (1618 in sqlite-utils alone) https://simonwillison.net/2026/Jul/25/ruff/25 Jul 2026 23:05 +0000at://did:plc:kft6lu4trxowqmter2b6vg6z/app.bsky.feed.post/3mriy5j5l322jhttps://bsky.app/profile/simonwillison.net/post/3mrbjg3u5tk2zI wrote about the completely wild incident where OpenAI were testing a new model and it broke out of its sandbox and broke INTO Hugging Face to steal the answers to the benchmark https://simonwillison.net/2026/Jul/22/openai-cyberattack/22 Jul 2026 23:53 +0000at://did:plc:kft6lu4trxowqmter2b6vg6z/app.bsky.feed.post/3mrbjg3u5tk2zhttps://bsky.app/profile/simonwillison.net/post/3mrakxzlam22fSan Francisco tip: it only costs around $15 ($10 in quarters plus a $5 bill for the self-playing violin) to activate every single Orchestrion in Musée Mécanique22 Jul 2026 14:48 +0000at://did:plc:kft6lu4trxowqmter2b6vg6z/app.bsky.feed.post/3mrakxzlam22fhttps://bsky.app/profile/simonwillison.net/post/3mr7bxnbqls2xToday at Pier 39 in San Francisco I learned that if you want to take a good photograph of a sea lion, you should aim to get their teeth22 Jul 2026 02:34 +0000at://did:plc:kft6lu4trxowqmter2b6vg6z/app.bsky.feed.post/3mr7bxnbqls2xhttps://bsky.app/profile/simonwillison.net/post/3mr5ucqec2s24I interviewed Cat Wu and Thariq Shihipar from the Claude Code team at the AI engineer World's Fair a couple of weeks ago - the video is now out, so I've published an annotated transcript of our conversation https://simonwillison.net/2026/Jul/21/cat-and-thariq/21 Jul 2026 12:57 +0000at://did:plc:kft6lu4trxowqmter2b6vg6z/app.bsky.feed.post/3mr5ucqec2s24https://bsky.app/profile/simonwillison.net/post/3mr4pqpbhb22yI got a fun photo last night of a California Brown Pelican surrounded by Elegant Terns21 Jul 2026 02:03 +0000at://did:plc:kft6lu4trxowqmter2b6vg6z/app.bsky.feed.post/3mr4pqpbhb22yhttps://bsky.app/profile/simonwillison.net/post/3mqxvkzieoc2tIf you have Claude Code installed you're running software that uses the new (unreleased) version of Bun that's been rewritten in Rust - here are two commands you can run to see that for yourself https://simonwillison.net/2026/Jul/19/claude-code-in-bun-in-rust/19 Jul 2026 04:04 +0000at://did:plc:kft6lu4trxowqmter2b6vg6z/app.bsky.feed.post/3mqxvkzieoc2thttps://bsky.app/profile/simonwillison.net/post/3mqspdej7wc2uSuggestion for hyperscalers feeling pressure over data center water use: Buy up a few exclusive country clubs, convert the golf courses into public parks, pay for guides and binoculars to get the previous members into birdwatching - help them embrace a more sustainable hobby!17 Jul 2026 02:29 +0000at://did:plc:kft6lu4trxowqmter2b6vg6z/app.bsky.feed.post/3mqspdej7wc2uhttps://bsky.app/profile/simonwillison.net/post/3mqs2y7gutk2aMy notes on Kimi K3, plus some thoughts on what we can still learn from the pelican benchmark even while it becomes further detached from how good the models are at the things that matter (like agentic tool calling across longer conversations) https://simonwillison.net/2026/Jul/16/kimi-k3/16 Jul 2026 20:25 +0000at://did:plc:kft6lu4trxowqmter2b6vg6z/app.bsky.feed.post/3mqs2y7gutk2ahttps://bsky.app/profile/simonwillison.net/post/3mqpwytj6lc2dI poked around in the just open sourced Grok Build CLI tool - 844,000 lines of Rust code! - and dug up a few interesting highlights, including a "self-contained terminal renderer for Mermaid diagrams" that renders them using Unicode box-art! https://simonwillison.net/2026/Jul/15/grok-build/16 Jul 2026 00:08 +0000at://did:plc:kft6lu4trxowqmter2b6vg6z/app.bsky.feed.post/3mqpwytj6lc2dhttps://bsky.app/profile/simonwillison.net/post/3mqnceqzbsc24Made myself a custom Codex pet today (you'll never guess what it was), and had some fun reverse-engineering how the feature works - it generates animated sprite sheets using gpt-image-2 https://simonwillison.net/2026/Jul/14/pedalican/14 Jul 2026 22:54 +0000at://did:plc:kft6lu4trxowqmter2b6vg6z/app.bsky.feed.post/3mqnceqzbsc24https://bsky.app/profile/simonwillison.net/post/3mqkzftto4226New TIL: Using uvx in GitHub Actions in a cache-friendly way I finally found a recipe that I like for running `uvx tool-name` in GitHub Actions without downloading a fresh copy of the package every time https://til.simonwillison.net/github-actions/uvx-github-actions-cache14 Jul 2026 01:08 +0000at://did:plc:kft6lu4trxowqmter2b6vg6z/app.bsky.feed.post/3mqkzftto4226https://bsky.app/profile/simonwillison.net/post/3mqfxbjx6f22oTrying to figure out the logistics of how you would go about nationalising Adele12 Jul 2026 00:46 +0000at://did:plc:kft6lu4trxowqmter2b6vg6z/app.bsky.feed.post/3mqfxbjx6f22ohttps://bsky.app/profile/simonwillison.net/post/3mqcn6ojgp22jThe Vergecast continues to be *so good* - I learn a bunch of new things every time I listen to it, and the quality of the commentary is universally excellent Just listened to the latest episode about Netflix content strategy and Meta’s smart glasses https://www.theverge.com/podcast/964082/netflix-youtube-smart-glasses-vergecast10 Jul 2026 17:08 +0000at://did:plc:kft6lu4trxowqmter2b6vg6z/app.bsky.feed.post/3mqcn6ojgp22jhttps://bsky.app/profile/simonwillison.net/post/3mqafoxnsl226Notes on GPT-5.6, which includes some interesting new additions to the API (programmatic tool calling and multi-agent in particular) - plus 18 pelicans for the 6 reasoning levels and 3 new models: https://simonwillison.net/2026/Jul/9/gpt-5-6/09 Jul 2026 19:48 +0000at://did:plc:kft6lu4trxowqmter2b6vg6z/app.bsky.feed.post/3mqafoxnsl226https://bsky.app/profile/simonwillison.net/post/3mq3e6p7gs227I released sqlite-utils 4.0, the 124th release but the first major version bump since 3.0 back in 2020 I managed to keep things backwards-compatible all the way up to version 3.39 before the accumulated design mistakes forced me to bump that number! https://simonwillison.net/2026/Jul/7/sqlite-utils-4/07 Jul 2026 19:38 +0000at://did:plc:kft6lu4trxowqmter2b6vg6z/app.bsky.feed.post/3mq3e6p7gs227