OpenAI models allegedly get 832 juice via API vs 128 in Codex
legit_api · x · 2026-09-28
A widely shared thread reports that the same OpenAI models appear to receive far more reasoning budget ("juice") through the API than inside OpenAI's own Codex harness — meaning even paying Max subscribers may be getting a constrained version. GPT-6 Astra is reportedly around 832 juice via API versus 128 in Codex. If the measurements hold, OpenAI is effectively holding back model capability on its own plans while direct API users get the full reasoning budget.
More from coding & agent
- OCR'd millions of pages of well data so AI agents can search old oil fields — oilmutt · 2026-09-28
- Muse called the 'most hackable AI OS' for spinning up Docker, tools and videos autonomously — NathanWilbanks_ · 2026-09-28
- DefaultBrowserChanger: open-source macOS tool isolates AI agent test sessions with one-click browser switching — sormagec · 2026-09-28
- Google Antigravity adds native inter-agent messaging and user-to-agent DMs — rseroter · 2026-09-28
- Dataviz expert says AI pushes him toward lower-level charting languages for finer control — randal_olson · 2026-09-28
- Cloudflare Birthday Week 2026: native Rust in Workers, agentic browser, EmDash CMS — ritakozlov · 2026-09-28