OpenAI Launches GPT-6 Sol and Luna With 50% Lower Prices as AI Industry Debates Development Slowdown
OpenAI has launched two new artificial intelligence models, GPT-6 Sol and GPT-6 Luna, expanding its latest-generation model lineup with a focus on affordability and efficiency. The company says the new models deliver improved accuracy while costing significantly less to run than their GPT-5.6 predecessors.
The launch comes shortly after OpenAI introduced GPT-6 Astra, its more advanced model aimed at demanding tasks. While Astra focuses on higher-end workloads, Sol and Luna are designed to make advanced AI capabilities more accessible for large-scale professional use, including coding, automation, computer-use tasks and routine knowledge work.
OpenAI says GPT-6 Sol makes around half as many mistakes as GPT-5.6 Sol on its internal factuality evaluation. The company also says the new models build on improvements introduced with Astra in areas including reasoning, alignment, factual accuracy, coding and computer use.
The biggest change, however, is pricing. GPT-6 Sol costs $2 per million input tokens and $10 per million output tokens, representing a 50% reduction compared with the promotional pricing of GPT-5.6 Sol. GPT-6 Luna is priced at $0.10 per million input tokens and $0.50 per million output tokens. OpenAI attributed the lower prices to improvements in inference and caching efficiency, saying the savings are being passed on to customers.
The two models serve different workloads. Sol is positioned for more demanding professional applications, including coding and complex tasks, while Luna is aimed at high-volume work where speed and cost are important, such as summarising documents, extracting information and handling routine queries.
The releases also arrive at a notable moment for the AI industry. Earlier in September, Anthropic CEO Dario Amodei called for companies to slow the pace of frontier AI development, arguing that additional time would allow the industry to strengthen safeguards and better understand increasingly capable systems. OpenAI CEO Sam Altman subsequently agreed that the pace of frontier development should be managed and said OpenAI would support independent evaluation of its systems.
Despite those calls for greater caution, OpenAI and Anthropic have continued releasing new models. Anthropic introduced Claude Opus 5.5 around the same time as OpenAI's GPT-6 Sol and Luna launch, adding to the increasingly competitive push toward models that combine stronger performance with lower operating costs.
The emphasis on cheaper inference reflects a broader shift in the AI market. As businesses increasingly integrate generative AI into software, coding tools and automated workflows, the cost of processing large volumes of requests has become an important consideration alongside raw model performance.
For OpenAI, the Sol and Luna releases therefore extend GPT-6 beyond the most demanding applications and into higher-volume workloads where lower token costs could make advanced AI easier to deploy at scale.
