Back to the stories

OpenAI rolls out GPT-6 Sol and GPT-6 Luna globally across ChatGPT tiers

Score 10.0,

Model

Building on our coverage from one day earlier, OpenAI is now rolling GPT-6 Sol and GPT-6 Luna across all ChatGPT tiers worldwide.

The rollout started by putting GPT-6 into the Chat tab for paid plans first, then expanding to the Free and Go tiers the next day. OpenAI says these new Sol and Luna releases replace the GPT-5.6 versions that powered general ChatGPT, while earlier GPT-6 builds continue to run inside developer tools like Codex and ChatGPT Work.

Why this matters: the assistant you and your teams use every day is getting the latest frontier-model behavior. OpenAI targets this update at the product that reaches people most directly, roughly 1.2 billion weekly ChatGPT users, so capability improvements stop being a research milestone and become an everyday feature.

How it works, simply: think of Sol and Luna as two tuned releases of the same engine, a higher-capability flavor for paid plans and a lighter flavor for free access. OpenAI also ships October-specific variants to ChatGPT while keeping separate versions for developer and work-focused products, like different release channels in ordinary software.

What changes now: everyday ChatGPT conversations should get noticeably better across tiers, and businesses that rely on ChatGPT will have those gains immediately. At the same time, developers using Codex or ChatGPT Work will not see identical behavior because those products use distinct variants. And this does not mean GPT-6 is suddenly runnable on a laptop, it is delivered through OpenAI's hosted service.

What's next: the key question is whether these product-side upgrades will shift who builds on which model, and how quickly independent benchmarks and developers can measure the difference.