OpenAI has launched GPT-6.1 Sol, a new model it says delivers nearly the same performance as GPT-6 Astra for agentic coding, computer use and professional work, while costing one-fifth as much for standard input and output tokens.
The model was introduced at OpenAI’s DevDay event, only a week after GPT-6 Sol. OpenAI said GPT-6.1 Sol improves on its predecessor in programming, debugging, document understanding and multistep workflows, with performance approaching GPT-6 Astra in several evaluations.
The company also reported better factual accuracy. At low reasoning effort, responses containing factual errors fell from 11.4% with GPT-6 Sol to 7.7% with GPT-6.1 Sol. Across all reasoning settings, OpenAI said the model’s error rate remained within 1.9% of GPT-6 Astra.
OpenAI said GPT-6.1 Sol is also more transparent about its limitations and more consistent in following user intent and safety constraints. The company is not launching the expected GPT-6.1 Astra.
GPT-6.1 Sol is available to Plus, Pro, Business, Enterprise and Edu users in ChatGPT Work and Codex. It is not yet available in Chat.
Source: TechCrunch
