OpenAI DevDay: dots agent, GPT 6.1 Sol, Pro Max, and Decisions API
OpenAI DevDay: dots agent, GPT 6.1 Sol, Pro Max, and Decisions API
At DevDay, OpenAI unveiled dots, an always-on agent with its own cloud computer, plus GPT 6.1 Sol, a Pro Max subscription, and a private Decisions API preview.
OpenAI DevDay: new models, agents, and subscriptions
The main announcement is dots, a competitor to Muse and Grok Bot. It is an always-on AI agent with its own cloud computer, and it does not use Codex limits. The feature is available for Pro, Business Premium, and Enterprise subscriptions.
GPT 6.1 Sol
OpenAI is accelerating its release cycle: GPT 6.1 Sol arrives just one week after GPT 6 Sol. The company promises performance close to Astra at token prices five times lower.
Pro Max subscription
OpenAI announced a Pro Max subscription priced at $500, offering 25 times the usage of the Plus plan. It includes an Ultrafast mode for Astra, which is 8 times faster and 6 times more expensive. Ultrafast is also available in the API, and OpenAI says it will add an Ultrafast mode for GPT 6.1 Sol.
Third-party app integration
OpenAI subscription limits can now be used in 16 third-party apps, including Devin, Notion, and OpenClaw, by signing in through ChatGPT. The company is expected to extend this support to more applications.
Decisions API
OpenAI reportedly built a competitor to Jev in just two weeks, based on Luna. Decisions API is currently available in private API preview, with a public launch promised in the coming days.
Why it matters
AnalysisThe announcements signal OpenAI’s push toward persistent cloud agents, faster model releases, and broader subscription integration across third-party apps.
Discuss in community
Share your questions and insights with developers
Related in Ecosystem
Practical guides, core terminology, and agent skills
Loop Engineering: Architecture of Autonomous Agentic Loops and the Post-Prompting Era
A deep dive into Loop Engineering: self-sustaining execution loops for Claude Code and autonomous agents, the DOER/CHECKER topology, deterministic verification, the /goal primitive, and reward hacking mitigation.
Claude Code /goal: Autonomous Agent Execution Without Human Intervention
Claude Code is a tool for working with code directly from the terminal. By default, it works in step-by-step mode: you write a request, get an answer, write again. The /goal command fundamentally changes this logic.
Short-Term and Long-Term Memory of Agents
The architecture of multi-tier memory for autonomous agents. It distinguishes between Short-Term Memory (active context window of the current session) and Long-Term Memory (vector databases, user profile files, and episodic diaries).
Telegram Bot API for Autonomous Agents
The official HTTP interface for the Telegram platform, enabling the creation of autonomous chatbots, AI assistants, Telegram Mini Apps (TMA), and interactive notification channels.