Theo - t3․gg
Summaries
I think I figured it out
Opus 5.5 is the model for daily work, and Fable 5.2 is the one to call when no objective test can judge the result. The subscription advantage holds, but the split is less tidy ...
Nobody Is Competing With Anthropic
Fable 5.2 is faster and cheaper than its predecessor, and it tops Anthropic's own benchmark tables, but in tests on real codebases it trails Opus 5.5 and Sonnet and needs a huma...
Gemini 4 Still Isn't What I Need (Yet)
Gemini 4 Carbon cuts the false-claim rate of its predecessor roughly in half, but it is too slow, too costly, and too error-prone to replace Opus for everyday coding. Its cleare...
I'm done with terminals
Terminals have stopped being the right place to run AI coding agents, and a graphical interface that tracks several agents across projects and machines serves that work better. ...
Dear Google, we need to talk.
Google is losing top talent and falling behind in AI agents because its culture blocks internal experimentation and punishes successful tools.
How I code with AI changed a lot
Five months after the previous video, the author abandons cursor plan mode and now relies on GPT‑5.5 for most coding tasks. Clerk provides unlimited apps on a $20 plan, handling...