
What happened
OpenAI doubled Work and Codex quotas on the Pro 100 and Pro 500 plans, and on Pro 200 doubled the lower quota while reversing a planned cut after user pushback.
Why it matters
Doubling a quota doesn't guarantee twice as many completed tasks, OpenAI says, since consumption varies by model, task type, and settings — so the usable gain may be smaller than the headline number suggests.
WHO IT HITSIndividual subscribers and teams on OpenAI's paid plans get more headroom for Work and Codex tasks, though actual throughput may not double because usage varies by model and settings.
Summaries like this, every morning.
OpenAI's decision to double usage limits on several paid tiers comes with a caveat: the company says a doubled quota does not translate directly into twice as many completed tasks, because consumption varies by model, task type, and settings. That gap between the headline number and real-world capacity is central to how customers should read the change.
The reversal on Pro 200 is particularly telling. OpenAI had planned a cut, but user pushback led the company to double the lower quota instead. That responsiveness suggests the company is weighing subscriber experience heavily as it adjusts plan structures.
Separately, OpenAI says it will keep shipping better models and efficiency gains to push usable capacity higher. It recently released its 6.1 models Terra and Sol, which it describes as much more efficient while matching or beating their predecessors. The expanded limits for the new "Dots" also stay in place for now, though Dot usage is tracked separately. The move comes as Anthropic prepares for its IPO, with both companies competing hard for B2B dominance in AI.
Pick your industry and the AI tools you use, and get news related to your work every day.
Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →
Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.
WeSales launched the free "Gem 移行チェック" tool, which asks three questions and shows what carries over, what does…

Aries Zhou released Howler, an open-source macOS app that summons a parchment-style letter with Option+H, writ…

In llm-d tests on 8 vLLM pods across 16 H100s, precise prefix-aware routing cut p90 time to first token from 9…

Serval raised a $150 million Series C at a $3 billion valuation, led by Sequoia, its third round in 12 months…

Meta quietly launched Style Sketch, an AI fashion app available in Indonesia, Singapore, and Thailand on iOS a…

Rivian is licensing elements of its AI voice assistant to Volkswagen Group within their $5.8 billion joint ven…
