← Back to feed

Special: everything OpenAI announced today at DevDay 2026

At DevDay 2026 on September 29 in San Francisco, OpenAI announced more than 20 launches: dots, always-on agents powered by GPT-6 Astra; GPT-6.1 Sol, near-Astra capability at one fifth of Astra's token prices ($2 input and $10 output per million in the API); Astra Ultrafast; Codex in the cloud; computer use in the Agents API; plugin extensions and new collaboration tools in ChatGPT.

Special: everything OpenAI announced today at DevDay 2026

OpenAI used DevDay 2026 in San Francisco on September 29 to make one argument: AI should stop waiting for the next prompt and start carrying ongoing work. The company announced more than 20 launches across ChatGPT, Codex, its models and the API. The headline pieces are always-on agents called dots, a cheaper near-frontier model called GPT-6.1 Sol, and a Codex that now runs in the cloud.

For developers, the practical change is cost and reach. Near-Astra capability now costs a fraction of Astra's price, agents can operate software through the API, and ChatGPT itself is opening up as a surface where third-party experiences run natively.

Dots: agents that keep working

Dots are OpenAI's new always-on agents. According to OpenAI, they are powered by GPT-6 Astra, have their own cloud computer and browser, learn from feedback over time, and can work toward goals around the clock.

Through OpenAI's plugin ecosystem, dots can connect to more than 4,000 apps. You can reach your dot in ChatGPT, Slack or Teams, and hop on a voice call when you want to talk something through.

OpenAI's own examples are concrete. A developer's dot watches customer feedback, scopes small fixes, builds and tests them, and brings back complete pull requests for review. A scientist's dot reruns analyses as new data arrives and updates the figures.

Control is built in. OpenAI describes access permissions and action review and approvals, and you can open your dot's computer at any time to inspect its work.

Availability: dots are rolling out on Pro and Business Premium in eligible markets. The first dot is included in those plans at no extra cost. Enterprise, Edu and Healthcare workspaces can try the beta when an admin enables it; it is off by default. OpenAI also previewed specialist dots for organizations, including support in Microsoft Agent 365.

GPT-6.1 Sol: near-Astra at one fifth of the price

GPT-6.1 Sol is an upgrade to GPT-6 Sol. OpenAI says it nearly matches GPT-6 Astra on agentic coding, computer use and professional work at one fifth of Astra's standard input and output token prices.

API pricing for gpt-6.1-sol is $2 per million input tokens, $0.10 per million cached input tokens and $10 per million output tokens. OpenAI says cached input is 95% cheaper than standard input and 50% cheaper than GPT-6 Sol's cached rate, which matters for agents that reuse long context.

On the benchmarks OpenAI published, GPT-6.1 Sol matches GPT-6 Astra on DeepSWE v1.1 at roughly one fifth of the cost and beats GPT-6 Sol's best score by 6.4 points. On OSWorld 2.0's offline set it comes within 2.1 points of Astra at roughly one seventh of the cost per task. On hard factuality prompts at low reasoning effort, the share of answers with a factual error falls from 11.4% to 7.7%.

OpenAI is also clear about limits: GPT-6 Astra still posts the top score on Terminal-Bench Science and remains the pick for the most difficult scientific work.

Availability: GPT-6.1 Sol is live in ChatGPT Work and Codex for Plus, Pro, Business, Enterprise and Edu users, and in the API. It is not yet available in Chat.

Speed and Codex

Ultrafast is OpenAI's new premium speed tier. It offers up to 8x faster token generation in Codex, around 300 tokens per second, and up to 6x in the API. GPT-6 Astra Ultrafast is available today in the API and in Codex on Pro 500 and Enterprise. GPT-6.1 Sol Ultrafast is coming soon.

Codex itself moves off the laptop. You can now run Codex on your own computer, remotely from a phone, or in the cloud from any device. Reusable development environments give teams a shared setup with approved settings and permissions.

The Codex CLI was refreshed. You can start and steer tasks by voice, and a new /agents view helps you delegate work and track several tasks at once. Editing prompts, resuming sessions and using worktrees also got easier.

Two review tools round it out. A new code review experience in the ChatGPT desktop app shows summaries and diffs and sends feedback to GitHub pull requests or GitLab merge requests; automatic reviews can take a first pass in the cloud. Codex Security Cloud scans entire GitHub repositories on demand or on a schedule, checks new commits, removes duplicate findings and prepares fixes.

A platform for agents

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, with OpenAI running the infrastructure.

With Amazon, OpenAI introduced Bedrock Managed Agents, powered by OpenAI. It takes the core capabilities of the Agents API and lets teams build OpenAI agents that run entirely in AWS.

A new Decisions API focuses Luna on user-defined questions with a fixed set of answers. Developers pass text or images and get back answers they can use to classify content, route requests or choose an agent's next action. It is in limited preview, with a broad release planned in the coming days.

ChatGPT is also opening up for builders. Plugin extensions let a plugin live in the sidebar, run interactive panels next to the conversation and add viewers for custom file types. A Plugin Creator and a redesigned submission flow aim to make publishing easier, and support for the proposed MCP Events specification lets plugins trigger automations when something changes in a connected app.

Working together, and the plans behind it

For teams, OpenAI introduced ChatGPT Space, a shared home where teammates, ChatGPT and your dot build on shared knowledge. Pages are a new document type for people and agents to edit together. Collaborative slides arrive in the coming weeks, with export to PowerPoint or Google Slides. Business and Enterprise teams can share pages and delegate recurring team tasks, and you can mention @ChatGPT in Slack and Microsoft Teams.

On subscriptions, Sign in with ChatGPT lets you use your plan allowance across 16 partner tools, including Cognition's Devin, Notion and Vercel. The new Pro 500 plan offers 25 times the ChatGPT Plus allowance and includes Ultrafast. For enterprises, the OpenAI Marketplace lets eligible customers apply part of an existing OpenAI commitment to approved partner software, starting with 32 partners. OpenAI Private Intelligence adds Zero Data Retention with Private Safety Processing, which runs automated safety reviews without giving OpenAI personnel access to the underlying content.

What to do

  1. If you run agents on the API, benchmark gpt-6.1-sol against your current model on your own tasks. The cached-input price is the number to model for long-context agents.
  2. If your team uses Codex, try cloud environments and the /agents view in the CLI, and point Codex Security Cloud at one repository to measure signal against noise.
  3. If you ship a ChatGPT plugin, read the plugin extensions and MCP Events docs and decide whether your product belongs in the sidebar rather than in a chat reply.

Many of these launches start on paid plans, and several are in preview or rolling out. Check availability for your plan and region before you build around them.

Watch the special

We summarized the announcements in two Spanish-language videos: a one-minute Short and the full two-minute special on Juan Chico's YouTube channel.

Sources

Related