OpenAI Introduces GPT-6.1 Sol: Near-Astra Intelligence at One-Fifth the Price

OpenAI has introduced GPT-6.1 Sol, a new model the company positions as offering near-Astra intelligence for coding, computer use and professional work.

Price is the central variable in the launch. OpenAI says GPT-6.1 Sol is priced at one-fifth of Astra's standard API input and output token rates.

Pairing near-Astra intelligence with a five-times-lower token price signals that Sol is aimed at usability under cost pressure rather than at the top of the capability stack. That is a trade-off built for the deployment stage.

The three use cases named in the release, coding, computer use and professional work, all involve long multi-turn tasks. Those workloads are highly sensitive to per-token pricing, because cost compounds with every turn and every token of context.

The five-times ratio also sets a reference point inside OpenAI's own lineup: Astra remains the flagship tier, while Sol sits below it and takes on the work that is called more often and tolerates more slack per task.

For developers, the real test is how much daylight sits behind the word near, and whether the cost advantage holds under realistic loads such as long context and tool calling.

The released summary offers positioning and pricing rather than benchmark detail, so more material is needed before the evaluation picture is complete. Teams choosing a model should still combine public evaluations with their own task tests to decide where Sol belongs in a pipeline.