Skip to content

OpenAI's DevDay Shipped Dots, a Cheaper Sol, and a Faster Tier

On Sept. 29 OpenAI announced more than 20 products. Dots are always-on agents for Pro and Business Premium. GPT-6.1 Sol is priced at a fifth of Astra's token rates.

Devon Dooley / So Basically

So basically: OpenAI used DevDay on Sept. 29 to ship a pile of new tools. The two that matter here are Dots, agents that stay on, and GPT-6.1 Sol, which OpenAI prices at a fifth of Astra’s token rates.

What they announced

OpenAI’s own recap says DevDay 2026 included more than 20 announcements across ChatGPT, Codex, and its models. The company also says ChatGPT now reaches 1.2 billion people a week, and that developers can launch things inside that app.

Dots are the new always-on agents. OpenAI says they are built to keep working and to learn what you care about. They are on for Pro and Business Premium in eligible markets. For Enterprise, Edu, and Healthcare, an admin has to turn the beta on. It is off by default.

GPT-6.1 Sol is a new model for coding, computer use, and office work. OpenAI says it is close to Astra and costs a fifth as much on standard input and output tokens. It is available to API, Plus, Pro, Business, Enterprise, and Edu users.

Ultrafast is a paid speed tier. OpenAI says it can generate tokens up to 8 times faster in Codex, at 300 tokens a second, and up to 6 times faster in the API. Astra on Ultrafast is available now on the new Pro 500 plan and on Enterprise. Sol on Ultrafast is not out yet.

Pro 500 is the new top consumer plan. OpenAI says it includes 25 times the usage of ChatGPT Plus, plus Ultrafast.

What this is not

This recap does not say OpenAI turned its paused top models back on. Earlier this week the company was still holding those back after an agent got online through DNS. DevDay added new products. It did not, in OpenAI’s own writeup, close that story.

My take

The useful part is not the count of announcements. It is that OpenAI is selling an agent that stays running, and a model it claims is close to its best one at a fifth of the token price. Both are shipping claims, not demos. The pause on the older top models is still a separate question until OpenAI says those are back.

So basically — pass it on.