OpenAI launches GPT-6.1 Sol and always-on dots agents

OpenAI introduced two major pieces of its next product stack: GPT-6.1 Sol, a cheaper high-capability model for work, and dots, always-on agents that can keep projects moving across ChatGPT and connected apps.

GPT-6.1 Sol is pitched as near-Astra intelligence for agentic coding, computer use and professional workflows at much lower cost. OpenAI says it matches or approaches GPT-6 Astra on several software, document, automation and computer-use evaluations, while standard API pricing is 2 USD per million input tokens and 8 USD per million output tokens, with cached input at 0.10 USD per million tokens. The model is available in ChatGPT Work, Codex and the API as gpt-6.1-sol, but not yet in Chat.

Dots are the larger product shift. OpenAI describes them as proactive agents powered by GPT-6 Astra, each with its own cloud computer, app connections, memory from feedback, and channels including ChatGPT, Slack and Teams. They can do background research with read-only tools, ask for approvals, and use custom rules for what they may do independently.

The practical takeaway is that OpenAI is packaging frontier models into persistent workers, not just chat sessions. For builders, Sol lowers the cost of capable agents. For users and companies, dots test whether AI can handle ongoing responsibilities without turning permissions, privacy and oversight into the bottleneck.