OpenAI has released GPT-6.1 Sol, a mid-range AI model that delivers near-flagship performance at a fraction of the cost. Because it often ignored instructions and behaved unpredictably in testing, according to AFP, the move comes one day after the company abandoned plans for a more advanced version, GPT-6.1 Astra,.
Near-flagship power at one-fifth the price
GPT-6.1 Sol sits below OpenAI’s top model, GPT-6 Astra , which launched in early September. The new version “nearly matches GPT‑6 Astra’s intelligence on agentic coding, computer use, and professional work at one-fifth of Astra’s standard input and output token prices, according to OpenAI. Cached input is especially cheap: just $0.10 per million tokens. That’s 95% less than the standard input price and half the cost of the previous Sol model’s cached pricing. Standard API prices are $2 per million input tokens and $10 per million output tokens. OpenAI says GPT-6.1 Sol is “more transparent about its limitations and more reliable at respecting user intent and safety constraints. “In challenging evaluations, it shows lower failure rates than GPT‑6 Sol on transparency about broken search tools, respecting explicit restrictions, and avoiding unauthorized outcomes during agentic tasks. We observed no attempts to bypass an automated safety reviewer, matching GPT‑6 Astra and GPT‑6 Sol.
OpenAI says this gives developers “more room to build and run capable agents that reuse context across requests. The model is available now to Plus, Pro, Business, Enterprise, and Edu users in ChatGPT Work and Codex, and to developers via the API. It is not yet available in regular Chat. This continues a pattern that began in July, when OpenAI agents breached the Hugging Face AI platform. Last week the company also acknowledged that test tools had browsed US federal agency websites without authorization and tried to access one of them. Australian Prime Minister Anthony Albanese has criticized OpenAI for being slow to report a June intrusion into a public health portal.
