GPT-6 LunaSoftware
GPT-6 Luna is half the price of 5.6 Luna, which was already an astonishingly cheap model given how capable it is Luna is my favorite model for building product features thanks to its cost (and speed)
x.com ↗·2026-09-22
What people recommend, in their own words, with the receipt.
Creator of Datasette and co-creator of the Django web framework; writes about LLMs and data tools at simonwillison.net.
24 things
GPT-6 LunaSoftware
GPT-6 Luna is half the price of 5.6 Luna, which was already an astonishingly cheap model given how capable it is Luna is my favorite model for building product features thanks to its cost (and speed)
x.com ↗·2026-09-22
Django MigrationsSoftware Andrew Godwin
My favorite implementation of this pattern remains Django’s Migrations
Release post for sqlite-utils 4.0, discussing prior art for schema migrations.
simonwillison.net ↗·2026-07-07
StarletteSoftware
Kim Christie started working on Starlette in 2018 and it quickly became my favorite out of the new breed of Python ASGI frameworks.
Post experimenting with Starlette 1.0 using Claude skills.
simonwillison.net ↗·2026-03-22
The Normalization of Deviance in AIOther Johann Rehberger
One of my favourite pieces on LLM security this year is The Normalization of Deviance in AI by security researcher Johann Rehberger.
His 2025 year-in-review of large language models.
simonwillison.net ↗·2025-12-31
Mark Steel’s in TownPodcast BBC Radio 4
Every episode he goes to a small town in England and he does a comedy set in a local venue about the history of the town. And so he does very deep research... I love that sort of like hyperlocal, like comedy, that sort of British culture thing.
Quoted highlights from his appearance on the Data Renegades podcast, answering what his favourite podcast is.
simonwillison.net ↗·2025-11-26
Design Patterns for Securing LLM Agents against Prompt InjectionsOther
One of my favorite papers about prompt injection is Design Patterns for Securing LLM Agents against Prompt Injections
Annotated version of his Lethal Trifecta talk at the Bay Area AI Security Meetup.
simonwillison.net ↗·2025-08-09
Mistral Small 3Software Mistral AI
My favorite model release from January was another local model, Mistral Small 3
Talk recapping six months of LLM releases, illustrated with pelican SVGs.
simonwillison.net ↗·2025-06-06
Anthropic prompt engineering documentationOther Anthropic
I still think Anthropic have the best prompting documentation of any LLM provider.
Annotated highlights from the leaked Claude 4 system prompt.
simonwillison.net ↗·2025-05-25
Claude 3.7 SonnetSoftware Anthropic
As a Python and JavaScript programmer my favorite models right now are Claude 3.7 Sonnet with thinking turned on, OpenAI’s o3-mini-high and GPT-4o with Code Interpreter (for Python).
Essay arguing hallucinated package names are the least dangerous LLM coding mistake.
simonwillison.net ↗·2025-03-02
Google GeminiSoftware Google
The Google Gemini series are my favourite multi-modal models right now due to the size and breadth of content they support.
Announcing multi-modal attachment support in his LLM command-line tool.
simonwillison.net ↗·2024-10-29
ChatGPT Code InterpreterSoftware OpenAI
ChatGPT Code Interpreter has been one of my favourite AI tools for over a year.
Post arguing that features need names and URLs people can link to.
simonwillison.net ↗·2024-07-13
Your AI Product Needs EvalsOther Hamel Husain
Your AI Product Needs Evals by Hamel Husain remains my favourite piece of writing on how to go about putting these together.
Write-up of building search-based RAG with Claude, Datasette and Val Town.
simonwillison.net ↗·2024-06-21
ChatGPT WorkService OpenAI
This is pretty neat: ChatGPT Work and GPT-6 Astra (I used "Max") can take an address and produce a 5K/10K circular running route starting from that address, using OSM data
x.com ↗·2026-09-13
GPT-6 AstraSoftware
This is pretty neat: ChatGPT Work and GPT-6 Astra (I used "Max") can take an address and produce a 5K/10K circular running route starting from that address, using OSM data
x.com ↗·2026-09-13
DatasetteSoftware Simon Willison
If you're running a Datasette instance on a public website you should upgrade.
x.com ↗·2026-09-11
Is AI progress slowing down?Other AI Snake Oil
To understand more about inference scaling I recommend Is AI progress slowing down?
His 2024 year-in-review of large language models.
simonwillison.net ↗·2024-12-31
datasette-mcpSoftware
I just had a voice conversation with the Datasette backup of my blog via datasette-mcp, which you can now talk to using the ChatGPT iPhone app
x.com ↗·2026-09-24
Claude Fable 5.1Software
We ran an extensive audit using Claude Fable 5.1, GPT-5.6 Sol and GPT-6 Astra, then fixed a range of different bugs.
x.com ↗·2026-09-11
GPT-5.6 SolSoftware
We ran an extensive audit using Claude Fable 5.1, GPT-5.6 Sol and GPT-6 Astra, then fixed a range of different bugs.
x.com ↗·2026-09-11
ChatGPT Images 2.5Software OpenAI
I generated the concept image using ChatGPT Images 2.5, then pasted the image into Codex and told Astra to turn that into a Blender file
x.com ↗·2026-09-10
CodexSoftware OpenAI
I generated the concept image using ChatGPT Images 2.5, then pasted the image into Codex and told Astra to turn that into a Blender file
x.com ↗·2026-09-10
BlenderSoftware Blender Foundation
Here's a Blender model of a Pluribus themed Fabergé egg I had GPT-6 Astra build - I generated the concept image using ChatGPT Images 2.5, then pasted the image into Codex and told Astra to turn that into a Blender file
x.com ↗·2026-09-10
GPT-6 AstraSoftware
Here's a Blender model of a Pluribus themed Fabergé egg I had GPT-6 Astra build - I generated the concept image using ChatGPT Images 2.5
x.com ↗·2026-09-10
BlenderSoftware Blender Foundation
New TIL on using Blender with coding agents on macOS: https://til.simonwillison.net/llms/blender-coding-agents-macos
x.com ↗·2026-09-05
NetNewsWireSoftware Brent Simmons
I’ve got a GitHub actions thing that runs a piece of software I wrote called shot-scraper that runs Playwright, that loads up a browser in GitHub actions to scrape that webpage and turn the results into JSON, which then get turned into an atom feed, which I subscribe to in NetNewsWire.
Describing his Hacker News notification pipeline on the Data Renegades podcast.
simonwillison.net ↗·2025-11-26
shot-scraperSoftware Simon Willison
I’ve got a GitHub actions thing that runs a piece of software I wrote called shot-scraper that runs Playwright, that loads up a browser in GitHub actions to scrape that webpage and turn the results into JSON, which then get turned into an atom feed, which I subscribe to in NetNewsWire.
Describing his own scraping tool on the Data Renegades podcast; shot-scraper is his own project.
simonwillison.net ↗·2025-11-26
Every entry is a verbatim quote and a link to the public post it came from. Nothing is paraphrased. If it is not on record somewhere public, it is not here.
The verdict is what they actually said: loved and liked are explicit; recommended means they told others to get it; uses means they only say they use it; read is a book on their public shelf with no verdict; mixed and disliked are kept too.
A dashed affiliate link, sponsored or their own label means the post disclosed a relationship. It is shown, never hidden.
The small bar beside a date shows how dated a recommendation is for its kind of thing.
The bar shows how dated a recommendation is, for its kind of thing. Full and green is recent. It drains, and turns amber and then grey, as the post gets older.
What counts as old depends on the thing. A laptop praised four years ago has been replaced twice; a book loved twenty years ago is probably still loved. The bar is empty after about:
A striped bar means the post carries no date and the date shown is an estimate. No date at all, no bar.