Nostr notes by Simon Willison

DeepSeek released a new OCR model - I got it working on an NVIDIA ...

2025-10-20T17:28:06Z

DeepSeek released a new OCR model - I got it working on an NVIDIA Spark (CUDA + ARM64) by starting a Docker container, running Claude Code as root with "claude --dangerously-skip-permissions" and telling it to figure it out

Took 4 prompts and 40 minutes https://simonwillison.net/2025/Oct/20/deepseek-ocr-claude-code/

Vibe coding is irresponsibly building software through dice ...

2025-10-07T15:05:23Z

Vibe coding is irresponsibly building software through dice rolls, not caring what code is produced

What about when engineers at the top of their game use AI tools responsibly to accelerate their work?

I propose "vibe engineering"! (with my tongue only partially in my cheek)

https://simonwillison.net/2025/Oct/7/vibe-engineering/

New on Niche Musems: my write-up of a visit to the Musical Museum ...

2025-09-21T15:58:38Z

New on Niche Musems: my write-up
of a visit to the Musical Museum in Brentford, London... player pianos, self-playing violins, and orchestrions! https://www.niche-museums.com/115

New release of my llm-openrouter plugin adds support for ...

2025-09-21T00:25:31Z

New release of my llm-openrouter plugin adds support for reasoning options and tool usage, which means you can now execute tools against 179 models made available via OpenRouter
https://simonwillison.net/2025/Sep/21/llm-openrouter/

Leaked memo from Amazon that warns existing H1B holders to avoid ...

2025-09-20T15:57:46Z

Leaked memo from Amazon that warns existing H1B holders to avoid travel back into the USA after the September 21st deadline - their lawyers evaluated the new executive order as not just affecting new applications https://www.businessinsider.com/read-memos-sent-big-tech-trump-h-1b-changes-2025-9

Posted some of my own notes on Shlok Khemani's (excellent and ...

2025-09-12T07:40:50Z

Posted some of my own notes on Shlok Khemani's (excellent and comprehensive) notes on how Claude and ChatGPT's memory implementations differ from each other https://simonwillison.net/2025/Sep/12/claude-memory/

Made two pelicans with the new Qwen3-Next-80B-A3B-Instruct and ...

2025-09-12T04:14:37Z

Made two pelicans with the new Qwen3-Next-80B-A3B-Instruct and Qwen3-Next-80B-A3B-Thinking models

🐧🦩 Who needs legs?!

https://simonwillison.net/2025/Sep/12/qwen3-next/

Some notes on the new Claude API web fetch tool, which I think ...

2025-09-10T17:28:35Z

Some notes on the new Claude API web fetch tool, which I think can be used safely despite the risk of prompt injection exfiltration attacks if you're really carful with the allowed_domains parameter https://simonwillison.net/2025/Sep/10/claude-web-fetch-tool/

Claude just released their new version of ChatGPT Code ...

2025-09-09T18:41:12Z

Claude just released their new version of ChatGPT Code Interpreter! It can run Python and Node.js code in a sandboxed Linux environment, and unlike OpenAI's version the Claude one can install additional packages

Here's my detailed review: https://simonwillison.net/2025/Sep/9/claude-code-interpreter/

I got Codex CLI and GPT-5 to help me modify the Transformers.js ...

2025-09-08T21:00:53Z

I got Codex CLI and GPT-5 to help me modify the Transformers.js Llama 3.2 chat demo to enable loading that 1.2GB model from a local folder instead of fetching it from a URL

Full details including the Codex transcript and prompts I used here: https://simonwillison.net/2025/Sep/8/webgpu-local-folder/

o3 and GPT-5 Thinking (not necessarily regular non-thinking ...

2025-09-06T22:48:31Z

In reply to nevent1q…9r2e
_________________________

o3 and GPT-5 Thinking (not necessarily regular non-thinking GPT-5) really do represent a step change in how effective models are at evaluating search results, so I'd like to see clarity from them on exactly which models they evaluated and when

The previously sensible advice to never use ChatGPT for search ...

2025-09-06T19:45:05Z

The previously sensible advice to never use ChatGPT for search needs to be rethought - GPT-5 in thinking mode is shockingly good at running searches now https://simonwillison.net/2025/Sep/6/research-goblin/

My notes on Kimi-K2-Instruct-0905, aka Kimi K-2.1 - an ...

2025-09-06T17:03:59Z

My notes on Kimi-K2-Instruct-0905, aka Kimi K-2.1 - an incremental improvement on Moonshot's previous trillion parameter open weights model, now with twice the context length (256k up from 128k) https://simonwillison.net/2025/Sep/6/kimi-k2-instruct-0905/

There is some good news here: the Hacker News thread about this ...

2025-08-25T09:47:07Z

In reply to nevent1q…9ln2
_________________________

There is some good news here: the Hacker News thread about this topic is refreshingly packed with people who clearly understand the nature and severity of the problem and get why none of the proposed mitigations are likely to work out https://news.ycombinator.com/item?id=45004846

Notes and a pelican for DeepSeek 3.1 ...

2025-08-22T22:42:53Z

Notes and a pelican for DeepSeek 3.1 https://simonwillison.net/2025/Aug/22/deepseek-31/

Notes on running the new Qwen-Image-Edit model on my Mac, using ...

2025-08-19T23:47:36Z

Notes on running the new Qwen-Image-Edit model on my Mac, using the prompt "Give the pelicans rainbow colored plumage" against a photo I took of some pelicans

It took almost ALL of my 64GB of RAM and needed 54GB of disk space https://simonwillison.net/2025/Aug/19/qwen-image-edit/

The new Gemma 3 270M open weights model from Google is really fun ...

2025-08-14T17:27:21Z

The new Gemma 3 270M open weights model from Google is really fun - it's absolutely tiny, just a 241MB download

I asked it for an SVG of a pelican riding a bicycle and it wrote me a delightful little poem instead

The real value should become apparent with fine-tuning - models this small will never be useful for general purpose tasks but could absolutely fly when fine-tuned for something specific

https://simonwillison.net/2025/Aug/14/gemma-3-270m/

GitHub Codespaces provides a free web-based development ...

2025-08-13T05:42:03Z

GitHub Codespaces provides a free web-based development environment backed by a Linux container... and it turns out it also includes a GITHUB_TOKEN environment variable that provides free API access to the GitHub Models collection of LLMs

... here it's running my LLM CLI utility https://simonwillison.net/2025/Aug/13/codespaces-llm/

Qwen3-4B-Thinking (a 4GB download) just became the first model ...

2025-08-11T00:16:26Z

Qwen3-4B-Thinking (a 4GB download) just became the first model I've tested to directly push back against my pelican riding a bicycle test, calling it "oddly specific and completely unrealistic" and saying "this request violates physics and biology"
https://simonwillison.net/2025/Aug/10/qwen3-4b/

One of the surprises for me from the GPT-5 launch yesterday is ...

2025-08-08T17:53:26Z

One of the surprises for me from the GPT-5 launch yesterday is how OpenAI removed access to older models (like GPT-4o) for most ChatGPT users at the same time as they rolled out the new model. I wrote about how that's been playing out here: https://simonwillison.net/2025/Aug/8/surprise-deprecation-of-gpt-4o/

New Gemini 2.5 Pro preview model today, released in advance of ...

2025-05-06T18:27:30Z

New Gemini 2.5 Pro preview model today, released in advance of Google I/O. Google claim it is state of the art in both frontend code generation and video understanding. https://simonwillison.net/2025/May/6/gemini-25-pro-preview/

if you're interested in AI ethics here are the notes I've ...

2025-05-05T02:35:43Z

In reply to nevent1q…fxw2
_________________________

if you're interested in AI ethics here are the notes I've collected so far https://simonwillison.net/tags/ai-ethics/

no need to be rude

2025-05-05T02:26:43Z

In reply to nevent1q…ddfe
_________________________

no need to be rude

that website is EXACTLY why this issue matters to me - they start ...

2025-05-04T23:23:25Z

In reply to nevent1q…vsq9
_________________________

that website is EXACTLY why this issue matters to me - they start the page with the claims that 10kg of coal is needed to produce 200 ChatGPT emails

Is that true? It had better be true for their campaign to make sense

I don't think it is

(For one thing, how is 10,000 ChatGPT "searches" using the same energy as composing 200 emails? What's a "search" in this case anyway?)

mainly still just experiments - here to justify using it ...

2025-05-03T01:01:17Z

In reply to nevent1q…g87l
_________________________

mainly still just experiments - here to justify using it frequently when the API Models are still cheap and a whole lot better

My favorite local model right now is a bit of surprise to me: ...

2025-05-02T23:45:12Z

My favorite local model right now is a bit of surprise to me: I'm really enjoying the relatively tiny Qwen3-8B, running the 4bit quantized version on my Mac using MLX

It's surprisingly capable given it's a 4.3GB download and uses just 4-5GB of RAM while it's running

https://simonwillison.net/2025/May/2/qwen3-8b/

I base my estimates of their efficiency on their API prices, ...

2025-04-30T00:17:27Z

In reply to nevent1q…jhk4
_________________________

I base my estimates of their efficiency on their API prices, which have almost all continued to drop - with the exception of o1 Pro and GPT-4.5 which are both diabolically expensive https://tools.simonwillison.net/llm-prices

where did you get the thing about OpenAI losing money on every ...

2025-04-30T00:12:38Z

In reply to nevent1q…37ud
_________________________

where did you get the thing about OpenAI losing money on every user?

I know Sam made a throwaway remark about that for the $200/month OpenAI Pro thing a few months ago but I pretty much wrote that off as hyperbole and a way of bragging about their "most powerful" (aka least efficient) models - which they've already replaced with the more efficient o3/o4-mini combo

I don't have any certainty at all over usage numbers, but ...

2025-04-30T00:00:30Z

In reply to nevent1q…uk4w
_________________________

I don't have any certainty at all over usage numbers, but I'm confident ChatGPT has more consumer users because anecdotally
most people haven't even heard of the competition (see also Google Trends)

I use DeepSeek as an example to illustrate that spending $100m on a training run may not actually be necessary after all, and optimistically they might be a lesson the other big AI labs learn

I'm genuinely not comfortable giving people advice on which ...

2025-04-29T23:54:31Z

In reply to nevent1q…7an4
_________________________

I'm genuinely not comfortable giving people advice on which LLM to use if they are environmentally conscientious because none of the companies release enough energy usage data for me to make a credible suggestion

I don't think I ever said I was pro-gold rush I have ...

2025-04-29T23:40:24Z

In reply to nevent1q…05kq
_________________________

I don't think I ever said I was pro-gold rush

I have complicated opinions about all of this! Hence my contradicting titles in attached screenshots from https://simonwillison.net/2024/Dec/31/llms-in-2024/#the-environmental-impact-got-better

The one opinion I'll stick by is that telling someone "every time you run a ChatGPT prompt you're using engi in mentally irresponsible amounts of energy" is misleading

(People really do say that)

given everything we've learned about the transformer ...

2025-04-29T23:28:56Z

In reply to nevent1q…zeyr
_________________________

given everything we've learned about the transformer architecture in the last five years my strong intuition is that if OpenAI hadn't figure it out someone else would have

it's using loads of energy and relieves. But it's also ...

2025-04-29T22:46:51Z

In reply to nevent1q…7rcz
_________________________

it's using loads of energy and relieves. But it's also serving (latest claim) 500 million weekly active users

they're losing money because that's how this stage in a ...

2025-04-29T22:42:09Z

In reply to nevent1q…mv2z
_________________________

they're losing money because that's how this stage in a gold rush works - everyone in the hot space raises a ton of money and burns through it trying to grow bigger than their competitors so that when they figure out the business model they can do it better than anyone else

Silicon Valley explained it pretty well https://www.youtube.com/watch?v=BzAdXyPYKQo&pp=0gcJCdgAo7VqN5tD

I feel like in that case it's pretty easy to argue that if ...

2025-04-29T19:01:33Z

In reply to nevent1q…37f4
_________________________

I feel like in that case it's pretty easy to argue that if just a few hundred people take home a pretty rock it will reduce the supply of pretty rocks by an amount that meaningfully impacts that particular environment

I don't see a few thousand or even a few million people using ChatGPT as comparable to that, assuming they also spend time watching Netflix / YouTube / scrolling social media media etc

the "but LLMs aren't actually useful" argument is ...

2025-04-29T16:31:21Z

In reply to nevent1q…2ptk
_________________________

the "but LLMs aren't actually useful" argument is covered by that piece too (it's very well constructed) https://andymasley.substack.com/p/a-cheat-sheet-for-conversations-about?#%C2%A7this-is-all-a-gimmick-anyway-why-not-just-use-google-chatgpt-doesnt-give-better-information

I'm glad somebody out there is brave enough to push back ...

2025-04-29T15:35:04Z

I'm glad somebody out there is brave enough to push back against the "personal ChatGPT usage is terrible for the environment" message https://andymasley.substack.com/p/a-cheat-sheet-for-conversations-about?utm_campaign=posts-open-in-app&triedRedirect=true

"If you want to prompt ChatGPT 40 times, you can just stop your shower 1 second early."

"If I choose not to take a flight to Europe, I save 3,500,000 ChatGPT searches. this is like stopping more than 7 people from searching ChatGPT for their entire lives."

I certainly won’t deny that there are an incredible new array ...

2024-08-31T14:07:47Z

In reply to nevent1q…7f79
_________________________

I certainly won’t deny that there are an incredible new array of footguns now available to anyone who wants them

I thought that too, but the more work I get done with LLMs myself ...

2024-08-31T13:44:14Z

In reply to nevent1q…ned2
_________________________

I thought that too, but the more work I get done with LLMs myself the less worried I am about that

I have a Go project I wrote from scratch in production now, despite not being remotely fluent in Go. It has comprehensive test coverage and even implements continuous integration and continuous deployment, which is why I’m confident it’s not a spectacularly bad idea

Would other people YOLO something like that to production without tests? Maybe, and that would definitely be a bad idea!

This meshes with my more optimistic take on AI-assisted ...

2024-08-31T13:03:27Z

In reply to nevent1q…g230
_________________________

This meshes with my more optimistic take on AI-assisted programming from last year: AI-enhanced development makes me more ambitious with my projects https://simonwillison.net/2023/Mar/27/ai-enhanced-development/

Forrest Brazeal: “I think that AI has killed, or is about to ...

2024-08-31T13:01:41Z

Forrest Brazeal:

“I think that AI has killed, or is about to kill, pretty much every single modifier we want to put in front of the word “developer.”

“.NET developer”? Meaningless. Copilot, Cursor, etc can get anyone conversant enough with .NET to be productive in an afternoon … as long as you’ve done enough other programming that you know what to prompt.”

From https://newsletter.goodtechthings.com/p/the-death-of-the-modified-developer
https://indieweb.social/@fatrat/113056013841232498

any chance it could handle tweet threads in the future? I have ...

2024-08-10T17:47:06Z

In reply to nevent1q…7sh9
_________________________

any chance it could handle tweet threads in the future?

I have some /epic/ threads I want to archive, 100+ tweets chained together, most of which are themselves quote-tweets

This one for example (only properly visible if you are signed in to Twitter) https://twitter.com/simonw/status/1077737871602110466

It turns out Google Chrome ships a default, hidden extension that ...

2024-07-09T17:54:05Z

It turns out Google Chrome ships a default, hidden extension that allows code on `*.google.com` access to private APIs, including your current CPU usage

You can test it out by pasting the following into your Chrome DevTools console on any Google page:

chrome.runtime.sendMessage(
"nkeimhogjdpnpccoofpliimaahmaaome",
{ method: "cpu.getInfo" },
(response) => {
console.log(JSON.stringify(response, null, 2));
},
);

More notes here: https://simonwillison.net/2024/Jul/9/hangout_servicesthunkjs/