Tell HN: Codex macOS app switches to Fast speed after update without asking
Sentiment Mix
Geography
Expert Signals
mfi
author • 1 mention
Hacker News
source • 1 mention
AI-Generated Claims
Generated from linked receipts; click sources for full context.
I just updated my Codex macOS app, which enables the new GPT-5.5 model.
Supported by 1 story
I've intentionally kept the speed to "Standard" to not burn through my tokens too fast.After the latest update of the app, I notices that OpenAI changed the speed to Fast without asking, a change that burns up to 1.5x more tokens.
Supported by 1 story
Related Events
GPT-5.5 Instant System Card
Uncategorized • 5/5/2026
OpenAI claims ChatGPT’s new default model hallucinates way less
LLMs • 5/5/2026
CopilotKit raises $27M to help devs deploy app-native AI agents
Uncategorized • 5/5/2026
Show HN: Orch8 – Durable workflow engine in Rust, one binary, Postgres or SQLite
Uncategorized • 5/5/2026
Show HN: Got tired of paying $100/mo for SEO tools, so I built an alternative
Uncategorized • 5/5/2026