OpenAI releases GPT-6.1 Sol: pricing, context and API details

OpenAI released GPT-6.1 Sol on 29 September. Its model documentation positions it for complex coding, computer use and professional work at a lower cost than Astra.

GPT-6.1 Sol accepts text and image input and produces text. The documented limits are 1,050,000 tokens of context and 128,000 output tokens. Tool calling requires the Responses API. Chat Completions is supported without tools. The model supports low through max reasoning effort, with medium as the default; none and minimal are unavailable.

Standard text pricing is $2.00 per million input tokens, $10.00 per million output tokens and $0.10 per million cached input tokens. These are base rates for requests with up to 272,000 input tokens. Larger prompts, processing modes and regional options can change the bill; consult the linked model page before budgeting.

For a migration, keep the existing model available while evaluating GPT-6.1 Sol on a small set of real tasks. Include a correct tool call, a deliberately invalid tool result, an image with ambiguous details and an answer that must conform to your application's schema. Measure accepted answers rather than assuming a larger context limit will improve accuracy.

Record both response time and total request cost. Long conversation histories can hide a cost increase even when the published token rate looks attractive. Set a maximum output budget, retain application-level validation and make failures visible to the operator. Promote the model only after the same task succeeds repeatedly with the intended endpoint and reasoning settings. This is our operational recommendation, rather than a claim that the model has passed a particular AZ Labs evaluation.