Codex Ultra Fast burns budget fast, highlighting a widening compute-class gap
herbiebradley · x · 2026-10-01
Miles Brundage and Herbie Bradley discuss OpenAI Codex's new Ultra Fast mode: blazingly quick but budget-hungry. Bradley notes running out of Codex budget used to be regular even on standard tiers, and argues most tasks are better served by parallelism than raw speed.
He frames AI compute as now split into three tiers — free, paid, and insiders at a handful of companies — with multi-agent workflows plus speed tiers widening the gap between the last two.
Related event: Codex Ultra Fast Mode Burns Through Budgets, Sparking Compute Divide Debate(4 posts)→
More from AGI Musings
- Gemini 4 Argon launches, agentic software factories and agent skills data headline daily reads — rseroter · 2026-10-01
- RSI today is really Recursive Technological Improvement, not self-improvement — voooooogel · 2026-10-01
- No signal on the Bay Bridge: AI's broad impact hinges on patchy internet access — soumitrashukla9 · 2026-10-01
- Are LLMs failing human benchmarks because they're superhuman? A cognitive science debate — xuanalogue · 2026-10-01
- AI Memory Could Be the Industry's Biggest Moat, and Portability Is the Real Fight — r0ck3t23 · 2026-10-01
- Alignment debate: prosaic techniques may not scale past ASI, researchers warn — JacquesThibs · 2026-10-01