Quote · Nerd Snipe with Theo and Ben
5 different models dropped last week & the GPT-5.6 usage limits are brutal
Where this was said
Open Letter to OpenAI: The Codex System Prompt Is Half Garbage
At 1:57:35 · chapter starts 1:48:20
The chapter opens with Theo's letter to 'Thibault and my friends at OpenAI' — a crash-out that has been building all episode. The core indictment is structural: OpenAI is copying Anthropic features but consistently grabbing the wrong parts and implementing them worse, eroding user trust without the excuse of attempting anything novel. The concrete example that arrives next is devastating: Theo has read the Codex system prompt in full, possibly the first person to do so recently, and found that approximately half of it is prescriptive frontend UI guidance dating from early 2025. It tells the model to use Lucid Icons specifically, to use Three.js for 3D elements, to 'provide updates every 30 seconds,' and to never end a session while tasks are running. When Theo feeds this prompt to GPT-5.6 Sol for a rating, it gives it 3 out of 10 as a general coding agent prompt and identifies the frontend section as 'a regression test suite, not guidance.' The same model produces dramatically better UIs in Claude Code, where the system prompt never mentions the words 'frontend' or 'UI' once.
Theo explained that going from High to X-High reasoning doubles token cost for a 1–2% benchmark improvement, and Max doubles it again, making X-High/Max 4× more expensive than High for marginal gains.
Ultra mode isn't a reasoning level — it's a system prompt append that forces Max reasoning AND spawns infinite recursive subagents, with no way to pass effort levels down the chain. Ben burned 50% of his weekly usage in a single 2-hour PR review. Don't touch it.
Codex Ultra mode not only triggers mass subagent spawning but forces every subagent to run at Max reasoning level, unlike Claude Code's UltraCode which pins subagents at High reasoning.
