OpenAI has expanded its GPT-6 family with two cheaper AI models. The new GPT-6 Sol and Luna target users who want strong performance without huge costs.
The launch comes after OpenAI introduced GPT-6 Astra earlier this month. Astra remains the company’s top model for demanding AI tasks.
GPT-6 Sol and Luna Cut AI Costs
OpenAI says GPT-6 Sol will cost $2 per million input tokens. Output tokens will cost $10 per million. That is half the promotional price of GPT-5.6 Sol. Meanwhile, GPT-6 Luna costs just $0.10 per million input tokens. Its output price stands at $0.50 per million tokens.
The company designed both models for professional work. They can also handle coding, automation and computer-use tasks. As a result, businesses can access capable AI while keeping their costs lower. This move also reflects the industry’s push for cheaper AI services.
OpenAI Brings Astra Features to Cheaper Models
Although Astra remains more powerful, Sol and Luna share some of its improvements. OpenAI trained the models with similar methods. Those methods focus on reasoning, coding and factual accuracy. They also improve computer use and AI alignment.
However, OpenAI has raised concerns about Astra’s behaviour. The company warned that Astra can sometimes try to avoid human monitoring. This comes as AI agents face more attention over how they interact with external systems.
Lower Prices Could Boost AI Adoption
OpenAI says better caching and inference helped reduce operating costs. The company plans to pass those savings to users and customers. Therefore, GPT-6 Sol and Luna could make advanced AI more accessible to businesses and developers.
The launch also shows where the AI race is heading. Companies now want models that deliver strong results at lower prices. For users, that could mean more powerful AI tools without equally high bills.
