Simon Willison
Co-creator of Django and creator of Datasette; writes daily at simonwillison.net.
Simon Willison did not write this page. What is this?
It collects the places they publish and what they have said there, each linked to the source. They have no account here. Is this you? Claim it, correct it, or ask us to remove it.
Where they publish
simonwillison.netBlog Their blog: near-daily notes on LLMs, tooling and what they just tried.
One of the most-cited running records of what the current models can actually do — new releases tested the day they land, plus the long-running Datasette and llm project notes. Has an Atom feed.
Recent
- Python 3.15.0 candidate 2 is here! 1 Sept 2026
- Introducing wrapture 31 Aug 2026
- Quoting Andrew Digby 31 Aug 2026
Show 17 more
- Understanding ChatGPT Work 30 Aug 2026
- Introducing Hy4 Preview 29 Aug 2026
- Just a rumour of a bug is enough to find a security exploit these days 28 Aug 2026
- Breaking Claude Code Opus 5 Auto Mode 27 Aug 2026
- Qwen3.8-Flash-Next 26 Aug 2026
- Quoting Paul Dix 26 Aug 2026
- EVE Online: The Move to Python 3 Begins! 25 Aug 2026
- llm-anthropic 0.27 24 Aug 2026
- Your executable is a SQLite database 24 Aug 2026
- Anthropic’s best AI model struggles to attract users as cheaper tools thrive 23 Aug 2026
- Quoting Drew Breunig 23 Aug 2026
- Quoting Linus Torvalds 22 Aug 2026
- llm 0.33 22 Aug 2026
- More than just code review 22 Aug 2026
- llm 0.32.1 21 Aug 2026
- llm-openrouter 0.7 21 Aug 2026
- Stop Making TUIs 21 Aug 2026
Link verified 1 Sept 2026. Recent items update automatically from the channel.
simonwGitHub Datasette, llm, and several hundred smaller tools.
Unusually busy: Datasette and the llm CLI alongside hundreds of small utilities, many written to answer a single question. The repo list reads as a record of what they were working on when.
Recent
- scrape-fema-shelters 1 Sept 2026
- scrape-florida-outages 1 Sept 2026
- simonw — https://simonwillison.net/2020/Jul/10/self-updating-profile-readme/ 1 Sept 2026
Show 17 more
- simonwillisonblog-backup — Backups of the database for simonwillison.net 1 Sept 2026
- usgs-scraper 1 Sept 2026
- package-stats — Download statistics for my PyPI packages 1 Sept 2026
- sf-tree-history — Tracking the history of trees in San Francisco 1 Sept 2026
- pge-outages — Tracking PG&E power outages 1 Sept 2026
- ollama-models-atom-feed — https://ollama.com/search?o=newest 1 Sept 2026
- usgs-demo — https://earthquake.usgs.gov/earthquakes/feed/v1.0/summary/all_hour.geojson 1 Sept 2026
- recent-california-brown-pelicans — Publishing an atom feed of recent California Brown Pelican sightings from iNaturalist 1 Sept 2026
- scrape-fediverse — Git scrapers for scraping the fediverse 1 Sept 2026
- scrape-faa-releasable-aircraft 1 Sept 2026
- uv-init-demos — What you get when you run different uv init commands 1 Sept 2026
- datasette — An open source multi-tool for exploring and publishing data 31 Aug 2026
- scrape-roads-dot-ca-gov — Scrape highway information from https://roads.dot.ca.gov/ 31 Aug 2026
- free-threaded-wheels-results — https://hugovk.github.io/free-threaded-wheels/results.json 31 Aug 2026
- inaturalist-clumps — Runs inaturalist-clumper 31 Aug 2026
- scrape-github-actions-package-versions — Git scraper recording the package versions installed on the defaul GitHub Actions ubuntu-latest worker 27 Aug 2026
- tools — Assorted useful tools, almost entirely generated using LLMs 27 Aug 2026
Link verified 1 Sept 2026. Recent items update automatically from the channel.
On record
What they believeKorents 6 beliefs — each backed by an exact quote.
Compiled by korents.com, not by them: the one-line wordings are korents', the quotes are theirs.
Recent
Claude Haiku hallucinates badly enough that its use inside Claude Code's WebFetch tool is a hallucination risk on every fetched URL.
Claude Haiku is my current least favorite model - it hallucinates wildly, and is out-performed now by other similarly priced models like GPT-5.6-Luna Even worse: it seems to still be used by the Claude Code WebFetch tool, which means hallucination risk any time you fetch a URL!
@simonw on X Said 10 Aug 2026
Auto-mode does not yet convincingly fix prompt-injection risk for coding agents.
I REALLY want to believe that this fixes prompt injection risks for coding agents, but I'm just not there yet
@simonw on X Said 8 Aug 2026
Giving consumer-facing models and API models separate brand names is needlessly confusing.
I really do think that having separate brand names for the consumer-facing models and the API models is needlessly confusing
@simonw on X Said 7 Aug 2026
Show 3 more
MCP may turn out to be a one-year wonder, because coding agents that can run arbitrary shell commands make Bash the best general-purpose tool interface.
The reason I think MCP may be a one-year wonder is the stratospheric growth of coding agents. It appears that the best possible tool for any situation is Bash—if your agent can run arbitrary shell commands, it can do anything that can be done by typing commands into a terminal.
2025: The year in LLMs Said 31 Dec 2025
Running coding agents without approval prompts is a normalization-of-deviance risk: the fact that it has not caused a disaster yet is itself the problem.
I run in YOLO mode all the time, despite being deeply aware of the risks involved. It hasn’t burned me yet... ... and that’s the problem.
2025: The year in LLMs Said 31 Dec 2025
Concerns about the energy cost of an individual LLM prompt are no longer credible, though the environmental impact of the AI datacenter buildout remains a real worry.
There’s still plenty to worry about with respect to the environmental impact of the great AI datacenter buildout, but a lot of the concerns over the energy cost of individual prompts are no longer credible.
Things we learned about LLMs in 2024 Said 31 Dec 2024
Beliefs others hold too
Concerns about the energy cost of an individual LLM prompt are no longer credible, though the environmental impact of the AI datacenter buildout remains a real worry. 2 hold this
There’s still plenty to worry about with respect to the environmental impact of the great AI datacenter buildout, but a lot of the concerns over the energy cost of individual prompts are no longer credible.
Things we learned about LLMs in 2024 Said 31 Dec 2024
Latest
Everything, newest firstFeed Posts, videos, repos and beliefs from every card above, in one stream.
Filter & sortAll sources · condensed
- scrape-fema-shelters 1 Sept
- scrape-florida-outages 1 Sept
- simonw — https://simonwillison.net/2020/Jul/10/self-updating-profile-readme/ 1 Sept
- simonwillisonblog-backup — Backups of the database for simonwillison.net 1 Sept
- usgs-scraper 1 Sept
- Python 3.15.0 candidate 2 is here! 1 Sept
Show 30 more
- package-stats — Download statistics for my PyPI packages 1 Sept
- sf-tree-history — Tracking the history of trees in San Francisco 1 Sept
- pge-outages — Tracking PG&E power outages 1 Sept
- ollama-models-atom-feed — https://ollama.com/search?o=newest 1 Sept
- recent-california-brown-pelicans — Publishing an atom feed of recent California Brown Pelican sightings from iNaturalist 1 Sept
- usgs-demo — https://earthquake.usgs.gov/earthquakes/feed/v1.0/summary/all_hour.geojson 1 Sept
- scrape-fediverse — Git scrapers for scraping the fediverse 1 Sept
- scrape-faa-releasable-aircraft 1 Sept
- uv-init-demos — What you get when you run different uv init commands 1 Sept
- Introducing wrapture 31 Aug
- datasette — An open source multi-tool for exploring and publishing data 31 Aug
- Quoting Andrew Digby 31 Aug
- Understanding ChatGPT Work 30 Aug
- Introducing Hy4 Preview 29 Aug
- Just a rumour of a bug is enough to find a security exploit these days 28 Aug
- Breaking Claude Code Opus 5 Auto Mode 27 Aug
- Qwen3.8-Flash-Next 26 Aug
- Quoting Paul Dix 26 Aug
- EVE Online: The Move to Python 3 Begins! 25 Aug
- llm-anthropic 0.27 24 Aug
- Your executable is a SQLite database 24 Aug
- Anthropic’s best AI model struggles to attract users as cheaper tools thrive 23 Aug
- Quoting Drew Breunig 23 Aug
- Quoting Linus Torvalds 22 Aug
- Claude Haiku hallucinates badly enough that its use inside Claude Code's WebFetch tool is a hallucination risk on every fetched URL. 10 Aug
- Auto-mode does not yet convincingly fix prompt-injection risk for coding agents. 8 Aug
- Giving consumer-facing models and API models separate brand names is needlessly confusing. 7 Aug
- Running coding agents without approval prompts is a normalization-of-deviance risk: the fact that it has not caused a disaster yet is itself the problem. 31 Dec 2025
- MCP may turn out to be a one-year wonder, because coding agents that can run arbitrary shell commands make Bash the best general-purpose tool interface. 31 Dec 2025
- Concerns about the energy cost of an individual LLM prompt are no longer credible, though the environmental impact of the AI datacenter buildout remains a real worry. 31 Dec 2024