GPT-6.1 Sol vs GPT-6 Astra: What are the key differences between OpenAI’s models? Check prices, features and more

OpenAI has launched GPT-6.1 Sol as a cost-effective alternative to its flagship model, GPT-6 Astra. It offers near-Astra performance for coding and professional workflows at a significantly lower API price, targeting developers and businesses needing strong capabilities without Astra's costs.

OpenAI on Tuesday added another model to its GPT-6 lineup with the launch of GPT-6.1 Sol, positioning it as a lower-cost alternative to its flagship GPT-6 Astra. The new model is designed to deliver near-Astra performance for coding, computer-use tasks and professional workflows, but at a significantly lower API price.

The launch comes shortly after OpenAI introduced GPT-6 Astra as its most capable model for demanding work. While Astra is aimed at complex reasoning, coding, research, computer use and document creation, OpenAI says GPT-6.1 Sol brings much of that capability to a cheaper model.

GPT-6.1 Sol vs GPT-6 Astra: What's the difference?

The biggest distinction between the two models is the balance between capability and cost.

GPT-6 Astra remains OpenAI's flagship model for its most demanding workloads. GPT-6.1 Sol, meanwhile, is targeted at developers and businesses that need strong performance across coding, computer use and professional tasks without paying Astra's full price.

OpenAI says GPT-6.1 Sol can approach Astra's performance on several evaluations. On DeepSWE v1.1, a benchmark for complex software-engineering tasks in real codebases, the company says Sol matches Astra at roughly one-fifth of the cost.

GPT-6.1 Sol vs GPT-6 Astra: Price

The API pricing difference is substantial.

GPT-6.1 Sol therefore costs one-fifth as much as Astra for standard input and output tokens. Its cached-input price is also substantially lower at $0.10 per million tokens, compared with $1 for Astra.

The lower cached-input price could be particularly relevant for AI agents that repeatedly reuse the same context across multiple requests.

Garvit Bhirani