OpenAI launches GPT-6.1 Sol, saying it nearly matches GPT-6 Astra and costs less

OpenAI has released GPT-6.1 Sol, a new model that the company says delivers significant improvements over GPT-6 Sol on complex professional work. According to TechCrunch, those gains cover code writing and debugging, document understanding, and the execution of multi-step business workflows.

The more interesting part is how OpenAI positions the model. The company says GPT-6.1 Sol comes close to the more capable GPT-6 Astra while costing less, which suggests buyers may no longer have to pay flagship prices for near-flagship capability.

Seen across the product line, GPT-6 Astra still marks the ceiling, while GPT-6.1 Sol tries to strike a more practical balance between capability and cost. For teams that call models at scale in production, a cheaper option that is good enough is often more attractive than a pure performance leader.

The release also fits a familiar pattern among model makers: alongside a flagship, ship a sub-flagship aimed at large-scale deployment, and use price and efficiency to win adoption.

Public details remain limited for now. The report does not include specific pricing, benchmark scores, or a general availability date, and it does not say which tasks show the largest gap between GPT-6.1 Sol and Astra.

What to watch next is whether developers and enterprise customers move workloads to GPT-6.1 Sol quickly, and how OpenAI draws a clear line between Sol and Astra in its lineup.