It is true!
I will be talking about writing for the web when you want to use more than words.
And why creative work is more important than ever!
Rich Harris, Paolo Ricciuti, Stanislav Khromov, and pngwn are coming to Ljubljana for Svelte’s 10th anniversary.
pngwn (opperpinguïn) presents: Page, unbound -- Tools for writing beyond text
November 19–20, 2026
sveltesummit.com/#speakers
3
129
Just released a new twinkleplop version with some significant performance improvements across both tokenisation and rendering!
I finally had time to go back and explore a few other ideas. I was also slightly annoyed that I wasn't faster on all sizes.
twinkleplop.pngwn.at/docs/be…
ALT A bar chart showing a 72% performance increase in twinkleplop version 0.2.2 over the previous version A smaller set of bar charts show language specific gains: - html, 68% - javascript, 57% - json, 77% - markdown, 121% - python, 45% - rust, 58% - sql, 76% - svelte, 84% - toml, 81% - tsx, 41% - typescript, 42%
After about a year, on and off, I am veeeery happy to release TWINKLEPLOP, a syntax highlighter of tremendous power!
The speed of prism with the customisation of Shiki.
1
7
292
Gradio Workflow on Product hunt!
Connect nodes to build AI pipelines, powered by Hugging Face producthunt.com/products/gra…
8
6
27
6,619
pngwn retweeted
Core maintainer Elliott Johnson dreams of streams and non-blocking, streaming UI with endless possibilities. Get a peek into the future of SvelteKit's server rendering capabilities at Svelte Summit.
sveltesummit.com
1
3
13
1,845
pngwn retweeted
What does it look like to take creative coding to the next level? From a blank canvas to production-ready visuals in a highly interactive fashion - this is what the Fragment toolkit promises. Raphaël Améaume tells us all about it at this year's Svelte Summit - join us!
sveltesummit.com
2
8
722
I would like to create new kinds of cheese. In a special cheese lab.
It will be me, a team of cheese scientists, some mice and Claude.
We can change cheese forever.
I will need 1 billion dollars.
Thank you.
This is the first result from our new molecular biology lab, where a team of Anthropic biologists is using Claude to explore and accelerate fundamental biology research. There, Claude works through data and literature to generate hypotheses and candidate biological systems to study. After our scientists review Claude’s hypotheses, they test the most promising ideas, with all lab work done by our scientists.
We’d like to extend this approach to a broad range of problems—in genomics and in other fields. If you have a proposal for a research question, we’d like to hear from you.
329
pngwn retweeted
I switched from Shiki to Twinkleplop in Animotion
• Bundle: 143 KB → 44 KB
• Package size: 11 MB → 180 KB
• Highlight speed: 11x faster (3.66 ms → 0.33 ms)
• Deleted ~100 lines of async loading code
After about a year, on and off, I am veeeery happy to release TWINKLEPLOP, a syntax highlighter of tremendous power!
The speed of prism with the customisation of Shiki.
6
15
277
17,502
This is why it took so long to get AGENTS.md
the right way to use model capabilities is not to ship 10x more features to prod
it's to spend more time understanding your users, trying experiments, building prototypes, learning about things you don't understand so that you can ship things that actually work
7
1,137
🌱 Been a long time since I updated my personal website, but now it's refreshed with a new look and a 100% PageSpeed score! It is of course built with Svelte and the Mochi framework - have a look! stanislav.garden/ #svelte #javascript
1
2
20
660
Added some docs so you can use twinkleplop with mdsvex!
mdsvex.pngwn.at/docs#with-tw…
After about a year, on and off, I am veeeery happy to release TWINKLEPLOP, a syntax highlighter of tremendous power!
The speed of prism with the customisation of Shiki.
1
4
525
After about a year, on and off, I am veeeery happy to release TWINKLEPLOP, a syntax highlighter of tremendous power!
The speed of prism with the customisation of Shiki.
12
45
587
72,105
How we did it: ML-Intern in HuggingChat ran it end to end.
It wrote 8,797 requests in 18 languages, labelled them with Qwen's OG 9B teacher, filtered, trained both students with TRL, evaluated on text and rendered images, and pushed it all to the Hub.
We just approved a budget and read the report!
1
2
11
2,085
Qwen-Image 2.1 draws its best pictures when a 9B "prompt rewriter" expands your request first. That is 20 GB in bf16 and uses 1700 words as system prompt, and thinks for ~1,600 tokens before answering.
We shrank it to 0.8B. It fits on a laptop now. 🧵
20
39
506
34,521
pAcInG tHe fRoNtIeR
Please welcome GPT-6 Sol and GPT-6 Luna to the GPT-6 universe.
GPT-6 Sol and Luna build on the advances behind GPT-6 Astra, bringing much of its strengths into faster and more affordable models to support work at scale.
We’ve also made caching and inference more efficient, and we’re passing the savings directly to you: 50% lower API prices for Sol and Luna compared with GPT‑5.6 promotional pricing.
1
3
394
I asked Claude what the name meant in an incognito chat and it basically just said that I have a reputation for being an idiot.
After about a year, on and off, I am veeeery happy to release TWINKLEPLOP, a syntax highlighter of tremendous power!
The speed of prism with the customisation of Shiki.
1
302
“As with our previous models it was tested by our mum and she thinks it’s just wonderful”
Opus 5.5 is our first model since we called for pacing the frontier. As with previous models, it was tested by external evaluators before release, including METR and Frontier Design.
On our most comprehensive alignment test, it achieves the strongest score to date.
2
4
453
It’s been like a week since they called for ‘pacing’. What is even happening.
Opus 5.5 is our first model since we called for pacing the frontier. As with previous models, it was tested by external evaluators before release, including METR and Frontier Design.
On our most comprehensive alignment test, it achieves the strongest score to date.
244