OpenAI DevDay 2026: public notes and recap posts

OpenAI wrote: “DevDay 2026 is our biggest yet, with more than 20 major announcements across ChatGPT, Codex, our models, and entirely new forms of working with AI.” The same post says the company opened ChatGPT “as a shared surface where humans and agents can collaborate and where developers can directly launch new native experiences to our collective 1.2B weekly users.”

The official recap lists the announcements below. Availability and plan limits are as stated on that page; some items were still rolling out after the event.

• Dots. “Dots are remarkably capable, always-on agents built to handle everything. They’re a whole new way to work with AI—one that gets to know what matters to you, is always working on your behalf, and takes important work off your plate so you get more of your time and attention back.”
• GPT-6.1 Sol. OpenAI described this as “a major upgrade to GPT-6 Sol with exceptionally strong performance on agentic coding, along with computer use, and professional work.” It “delivers near-Astra intelligence to everyone at a fifth of its standard input and output token prices.”
• Ultrafast. “Ultrafast is our premium speed tier for workloads where speed matters most. Ultrafast offers up to 8× faster token generation (300 tokens per second) in Codex and up to 6x in the API.” The recap says GPT-6 Astra Ultrafast was available that day in the API and in ChatGPT Work and Codex on Pro 500 and Enterprise plans, with GPT-6.1 Sol Ultrafast “coming soon.”
• Decisions API. “Decisions API enables real-time decision-making by focusing Luna's intelligence on a specific set of user-defined questions with finite pre-defined answers.” It was “available in limited preview” that day, with a broader release “planned in the coming days.”
• Agents API with computer use. The Agents API “now supports computer use so developers can build agents that interact with software to complete tasks.” It also “brings Codex’s multi-agent capabilities, tool search, tool calling, and context compaction into your application.”
• Bedrock Managed Agents, powered by OpenAI. OpenAI wrote that it worked with Amazon so customers can “build OpenAI agents that run entirely in AWS.”
• Plugin extensions. “Plugin extensions let you give your plugin a home in the sidebar and build interactive panels where people can work alongside the conversation.”
• ChatGPT Space, Pages, and collaborative slides. Space is described as “a new home for your team to collaborate with AI.” Pages are “a new type of document, built for human and agent collaboration.” Collaborative slides let “multiple teammates and agents… edit the deck at the same time.”
• Sign in with ChatGPT. “Use your allowance across 16 partners including Cognition’s Devin, Notion, Vercel, T3, OpenClaw, and Dactyl, and control how much each can use.”
• A new Pro tier. “The new Pro 500 plan offers our highest usage allowance at 25 times the ChatGPT Plus allowance and includes access to Ultrafast.”
• OpenAI Marketplace. “Eligible enterprise customers can apply part of their existing OpenAI commitment toward approved partner software.” The recap named a first group of 32 partners.

The same recap also lists Private Intelligence, Codex in the cloud, a refreshed Codex CLI, improved plugin creation and discovery, Sites hosting plugins, MCP events for plugin automations, team tasks, @ChatGPT in Slack and Microsoft Teams, a Meetings plugin, and shareable profiles. The official sessions page points to available session recordings.

The Help Center states that ChatGPT Pro “now offers Pro 500, a new $500/month plan that includes GPT-6 Astra Ultrafast and GPT-6.1 Sol Ultrafast.” It lists monthly prices as Pro 100 at $100, Pro 200 at $200, and Pro 500 at $500, and says Ultrafast “is available only on Pro 500” among Pro plans.

CNBC reported live from San Francisco. After the keynote, it wrote that the company “showcased more than 20 different products and features to the roughly 2,500 people in attendance.”

It quoted Sam Altman at the close of the keynote: “AI should be about giving people more power over their own lives, more tools to create, learn, discover, expand knowledge, and with everything we launch today, and the incredible creative spirit that you all bring, we hope that we’re going to get to that future. That is the future we are excited to build with you all.”

On Dots, CNBC quoted Altman: “Dots are remarkably capable, always-on agents built to handle really anything you can think of.” And: “I’ve been really surprised at what a difference this makes to my life. I feel like I’ve finally gotten some of my attention back.” The same report said Dots “have their own cloud computer, can connect to more than 4,000 apps and learn from feedback over time.”

On Private Intelligence, CNBC quoted Altman: “[Zero Data Retention] with private safety processing delivers frontier model safety without storing your content on our servers at all, and with private inference, we can even provide privacy at the time of inference.”

CNBC also reported a protest outside the venue, and captioned photographs of a Dots sticker pop-up “at Fort Mason on Sept. 29, 2026 in San Francisco.”

In a keynote-day interview, CFO Sarah Friar told CNBC, of an earlier price tier: “When we launched our $200 SKU, people thought we’d lost our minds, [that] no one’s ever going to spend that per month.” She added: “What we see is not only are people willing to spend more, they’re actually moving more towards consumption in consumer.”

The Verge wrote that DevDay was held “on September 29th in San Francisco,” and that the keynote “revealed several notable updates, including Dots,” GPT-6.1 Sol, “1.2 billion users per week,” and “a new ChatGPT Pro plan that costs $500 per month.” It also wrote that, unlike Meta’s Muse, “Dots will initially be available to paid subscribers of ChatGPT Pro, Business Premium, and Enterprise.”

Reis Sirdas wrote a first-person account of being at the event. “OpenAI DevDay 2026 was quite the spectacle. Things kicked off with a protest outside the venue entrance, followed by food served under canopies that were dripping water onto our heads, multiple live demo fails, and a lot of showboating.” An attendee told him “last year’s sessions were more technical.” Another told him “the food was much better in the past.”

On trying the new tools that day: “Ironically, Dots were to be found everywhere (all over the walls, a giant clock, stickers, even Dippin’ Dots ice cream), everywhere but my computer.” He wrote that he later got the Cloud agent working “somewhat,” signed up for a mobile beta to fix Codex Remote, and “found Dots on my account days later.” After the event, “Sol 6.1… has actually become my go-to coding model.”

He also wrote: “Almost none of the things they announced were available on DevDay.”

Ryan Mish wrote that he “arrived early at Fort Mason in San Francisco, excited to spend the day with 2,500 other people who build things with AI.”

After a red-eye home, he asked his new Dot “to check my email and Slack.” “It found an urgent client requirement, opened my Google Drive, filled out the required form, and had me approve it before sending an email reply that it was ready.”

A week later, he wrote: “Dots are an incredible idea, with execution that feels rushed.” He kept using his Dot “for accountability, transcripts, and to-dos, while managing development directly in Codex.” GPT-6.1 Sol was “a clearer success”: “it’s a good model that isn’t as capable as Astra, but lets me get through much more of my regular work.”

Of the sessions, he called “Building with Loops: What Comes After Agents?” “the best of the day for me.” The presenters, he wrote, “suggested making Sol the main agent you work with, while letting it pass simpler tasks with clear plans to Luna and bring in Astra for difficult questions or continuing review.” He also described a producer/consumer pairing: one agent builds, another “opens it in a browser as the intended user.” “That’s changed my practice.”

Rico Tan, writing from San Francisco, called the post “a firsthand OpenAI DevDay 2026 field report.” His summary: “Seeing Dots in person made the shift clear: AI is moving from answering questions to owning ongoing work. That makes boundaries, accountability, and human oversight more important than ever.”

He wrote that the operating question had become: “What work should it be allowed to own, what can it access, and how will I know when it gets something wrong?” The post also lists safeguards described for Dots, including a separate cloud computer, app-level permissions, Custom Rules, an Activity View, and “mandatory human handoffs for some sensitive steps,” and notes OpenAI’s own statement that Dots can still make mistakes.

Digital Meld wrote that co-founder Brad Groux attended DevDay on 29 September, and thanked OpenAI “for flying Brad to San Francisco.” “He spent September 28–30 meeting the OpenAI team and other attendees, going to sessions, and catching up with people over coffee and drinks.”

On the drive home from the airport, “Brad… called AVA, his dot. Before he got home, he had planned the next day and moved work forward across several projects.” “It rang like a regular phone call.” The post lists the sessions he attended, including the opening keynote, “What Comes After Loops?: Deep Dive Into Agentic Architecture,” “Harness-as-a-Service: Building with the Agents API,” and the closing Q&A with Sam Altman and Thibault Sottiaux, after which “he headed to the reception at Fort Mason.”

The company also published a set of 26 public trip updates.

“I attended OpenAI DevDay in San Francisco. The announcements included ChatGPT Space, Dots, Ultrafast, and updates to Codex. Here’s what I’m taking away:”

On the day of the event, elvis posted: “So back! Let’s connect if you are at DevDay.” After the Dots announcement: “I have been testing it for the last few days and it’s something special. Proactive capabilities are really great. I used it to organize my entire DevDay from travels to session/meetings setup. It has its own computer.” Later he called a shout-out to Pi “Highlight of DevDay for me,” and wrote a follow-up: “Sharing some notes from OpenAI DevDay in parts.” The first part asked how to optimize for cost and performance, and named “4 of the most important levers you absolutely must try,” beginning with “prompt.”

Marina Mogilko wrote that she asked Sam Altman at DevDay: “What should the founders build today so that the next generation of models doesn’t make it obsolete?” She quoted his answer as beginning: “Contrary to everyone’s fears, models have NOT been taking over. The rate of new successful startups being created…”

In another post from the day, she wrote: “We asked Sam and Tibo what they’d never delegate to AI. Both said: raising kids. Sam said his young kids get no screen time.” And: “I want AI to free up more time with my kids. That’s the part I want to keep.”

The next day: “Yesterday, I was at OpenAI DevDay, where Sam Altman introduced Dots - OpenAI’s new ‘always-on’ AI agents. And I think it gives us a pretty good idea of where productivity is heading. Today, even when we use AI every day, we still have to initiate most things.”

“My notes from @OpenAI DevDay.” The post is a photo of handwritten notes from the day.