I ran their "Write a POV-Ray file that shows a pelican riding on a bicycle" prompt against the new gpt-5-codex-mini and got the worst result yet! https://simonwillison.net/2025/Nov/9/pelican-on-a-bike-raytracer-edition/

For comparison, here are the pelicans riding bicycles drawn by GPT-5-Codex-Mini (the new model), GPT-5-Codex and full GPT-5 - all produced via the same hacked version of the Codex CLI tool

I also recorded a 7 minute YouTube video showing how I got Codex to reverse-engineer and then extend itself in order to draw me that pelican https://www.youtube.com/watch?v=9o1_DL9uNlM

Since there's no API access yet I got OpenAI's Codex coding agent to rewrite itself (in Rust) to add a new "codex prompt ..." command which I could use to run prompts against the private models that are only available within that tool - full details here […]

uv makes testing different projects against upgraded dependencies so much easier - no need to think about virtual environments, uv handles them almost invisibly

I wrote more about my uv testing tricks in this TIL https://til.simonwillison.net/python/uv-tests

And here's an example of one of my code research prompts

Here's my research repo - each of the 13 folders is a different research project, and the README is automatically updated by an LLM to include summaries describing each one https://github.com/simonw/research?tab=readme-ov-file#research-projects-carried-out-by-ai-tools

It's a great TikTok! https://www.tiktok.com/@belligerentbarbies/video/7568380008633257271

And in case you don't make it as far as the "miscellaneous tips" section, here's a bunch of lessons I learned about working with coding agents that I picked up along the way https://simonwillison.net/2025/Nov/4/datasette-10a20/#miscellaneous-tips-i-picked-up-along-the-way

Here's the September edition, if you want to get a preview and catch up with some news that's over a month old now https://gist.github.com/simonw/d6d4d86afc0d76767c63f23fc5137030

And it's not just Cursor... rival agentic coding IDE Windsurf announced their own custom RL-trained fast coding model today as well!

Here are notes and a pelican on Windsurf's new SWE-1.5 model https://simonwillison.net/2025/Oct/29/swe-15/

So that's DeepSeek, Qwen, Moonshot (Kimi), Z.ai (GLM) and now MiniMax all turning out frontier open weight models in China

The ai-in-china tag on my blog is pretty busy this year! https://simonwillison.net/tags/ai-in-china/