GPT-5.6 Is Last in 5.x Line, GPT-6 Coming in About a Month

OpenAI just told the world that GPT-5.6 is it. No 5.7, no quiet 5.8 patch six months from now. The company is closing the book on the 5.x line and moving straight to GPT-6, which sources say arrives in roughly a month.

What makes this jump interesting isn't just the version number. GPT-6 is reportedly built on a substantially larger pretrain than anything in the 5.x family, a break from the pattern of incremental fine-tunes and safety tweaks that defined most of this year's releases. Bigger pretrains cost more, take longer, and carry real risk if the scaling gains don't show up. OpenAI is betting they will.

The timing lines up with pressure from Anthropic, whose Fable 5 model has apparently been closing the gap, with Fable 5.1 already teased as the follow-up. For most of 2024 and 2025, OpenAI held a comfortable lead in raw capability. That comfort seems to be eroding, and a compressed one-month gap between 5.6 and 6.0 reads like a company that wants to reset the conversation before a competitor gets there first.

There's also a simpler explanation buried in the decision to retire 5.x entirely rather than milk it further. Naming a model line dead signals confidence that the next one is a real step up, not a rebrand. OpenAI has been burned before by shipping updates that felt more like maintenance than progress. Calling GPT-6 the end of the incremental era is as much a marketing move as a technical one.

What happens next month will tell us whether the bigger pretrain actually translates into a capability jump big enough to justify skipping straight past 5.7 and 5.8. If it doesn't, this will look like OpenAI panicking over Anthropic's momentum. If it does, GPT-5.6 will be remembered as the last stop before the real reset.