OpenAI Expands GPT-6 Lineup With Cheaper Sol and Luna Models
The launch comes as AI companies increasingly focus on lowering the cost of running powerful models while competition intensifies across the industry
OpenAI has expanded its GPT-6 lineup with two lower-cost models, GPT-6 Sol and GPT-6 Luna, giving developers and businesses more affordable options for professional work, coding, automation and computer-use tasks.
The models were introduced on Tuesday, following the launch of GPT-6 Astra earlier this month. OpenAI describes Astra as its most capable broadly deployed model, while Sol and Luna are positioned as less expensive alternatives that offer some of Astra’s capabilities.
GPT-6 Sol is priced at $2 per million input tokens and $10 per million output tokens. Those rates are 50% lower than the promotional prices previously offered for GPT-5.6 Sol. GPT-6 Luna is priced at just $0.10 per million input tokens and $0.50 per million output tokens, making it the lower-cost option within the new lineup.
OpenAI said Sol and Luna were trained using methods similar to those used for Astra, with improvements in reasoning, factual reliability, coding, computer use and alignment. The company said improvements in caching and inference allow the models to be served at lower costs, with those savings passed on to customers.
The pricing strategy reflects a broader shift in the artificial intelligence industry as companies seek to make increasingly capable models cheaper to operate. Lower inference costs can make it more practical for businesses to run AI agents repeatedly for software development, data processing, automation and other high-volume tasks.
The new models also expand the range of choices available to OpenAI customers. Astra remains aimed at demanding projects where maximum capability is the priority, while Sol and Luna provide alternatives for workloads where cost and scale are more important.
The launch comes as OpenAI faces growing competition from companies including Anthropic and developers of Chinese AI models. The company has also been reducing prices across its existing GPT-5.6 lineup, including an 80% cut to GPT-5.6 Luna pricing announced in July and a more than 20% temporary reduction for GPT-5.6 Sol in August.
For developers and businesses, the expansion of the GPT-6 lineup signals a growing emphasis on balancing AI capability with operating costs as the technology moves deeper into everyday software and enterprise workflows.
