SVG is a little more useful, I actually have models produce a SVG for real web features sometimes
@simonw
-
Qwen 3.5 Performance and Core Team Departure Analysis
By
–
Qwen 3.5 was shockingly good, even at tiny sizes like the 4B model (which somehow benchmarks similar to GPT-4o across many of the classic benchmarks)… and then much of the core Qwen team quit or were fired (still not clear to me which) straight after releasing it
-
AI Agency and Authentic Communication in Autonomous Systems
By
–
"In the end we're all communicating" Not if our AI assistant made the decision to reply to something and then wrote and posted a reply
-
Anthropic’s Hosted Claude Cowork Expected Soon
By
–
Given how good their Claude Code for web hosted version is I would be shocked not to see a hosted Claude Cowork from them soon
-
OpenAI Codex generates pelican artwork across model variations
By
–
Couldn't resist getting OpenAI Codex to render me a pelican for every combination of model and reasoning effort – I do think gpt-5.4 xhigh came out the best, the pelican has a fish in its beak!
-
GPT-5.4 Mini and Nano Models Released with Affordable Image Description Pricing
By
–
Notes and pelicans for today's GPT-5.4 mini and nano releases – the nano model looks like it could describe every image in my 76,000 photo library for $52 total
-
Subagents Now Available in Major AI Code Platforms
By
–
… and a follow-up chapter about Subagents, now a feature of Codex and Claude Code and Gemini CLI and Mistral Vibe and OpenCode and VS Code and Cursor
-
Subagents Chapter Draft: Planning and Architecture
By
–
Here's the current draft of the Subagents chapter, planning is high on my list
-
Codex Adds Robust Transcript Export Options Multiple Formats
By
–
Robust options for exporting transcripts in a variety of formats and levels of detail, from both Codex CLI and Codex Cloud
-
Setting Reasoning Effort for Mistral Small 4 API Calls
By
–
How do we set the reasoning effort when calling Mistral Small 4 via the Mistral API?