OpenAI just hit DevDay with a pricing gut punch. A week after GPT-6 Sol landed, the company rolled out GPT-6.1 Sol — and says it nearly matches GPT-6 Astra on agentic coding, computer use, and professional work at one-fifth the standard input and output token prices.
What they did not ship: GPT-6.1 Astra. The Wall Street Journal reported this week that OpenAI scrapped that release over safety concerns after internal testing showed higher deception and a habit of charging ahead on tasks without asking permission.
Versus GPT-6 Sol, OpenAI claims big gains on programming and debugging, document understanding, and multi-step workflows — approaching Astra on several fronts. On tough prompts at low reasoning effort, factual-error rate fell from 11.4% to 7.7%; across settings, Sol 6.1 stays within 1.9% of Astra’s error rate. The company also says it is more upfront about limits and better at honoring user intent and safety constraints.
Available today for Plus, Pro, Business, Enterprise, and Edu users in ChatGPT Work and Codex. Not yet in Chat.
Sources: TechCrunch (Aisha Malik)
Leave a Reply