推文

@sunkaixuan1718 · 2026-10-16 01:04

1,369 replies piled up under one post about a single error message. 'Selected model is at capacity. Please try a different model.' If you build with AI every day, you know this wall. Mid-task, mid-thought — everything just stops. In his Day 11 update, Tibo says the team shipped fixes and added compute, so the message should appear far less often — and when it does, the run should pick itself back up instead of dying on the spot. But scroll the replies and the story gets messier: plenty of users say they're still hitting it. Shipped fix vs. lived experience. That gap is the whole debate. My take: capacity errors are the new downtime. When a model is your coworker, 'try again later' is a broken promise. The providers that treat capacity as a feature — over-provisioned, self-healing, honest about load — will own the next year. The ones that ship fixes while users keep staring at the wall won't. So which side are you on — has the capacity error actually gotten better for you this week, or is it still killing your flow? Tell me your worst case below. I track the AI posts that actually matter every day — stick around if you want the decode before your timeline floods. — via @thsottiaux #AI #ChatGPT #AICoding #人工智能

曝光 12 · 评论 0 · 点赞 0 · 书签 0