OpenAI and Anthropic announced lower-cost artificial intelligence models on Tuesday, expanding their flagship model families as competition intensifies across the global AI market. The launches target customers seeking stronger performance while limiting computing expenses, particularly as open-weight alternatives gain traction among businesses, developers, and research organizations worldwide today and startups.
OpenAI introduced GPT-6 Sol and GPT-6 Luna, saying their application programming interface prices are 50% below promotional pricing for GPT-5.6. Sol sits below the Astra tier and is designed for complex work such as coding, while Luna focuses on high-volume activities, including document summarization and information extraction tasks for enterprises.
Anthropic separately released Claude Opus 5.5, the newest model in its Claude 5.5 family. The company said it is more token-efficient than Opus 5 and costs about 40% less to operate. Anthropic product executive Dianne Penn said the company is working to reduce token use according to users’ selected effort.
These releases are the first from either company since concerns about advanced AI safety intensified. Former Anthropic researcher Jacob Coxon contributed to the debate after resigning and warning that leading laboratories were taking serious risks. Anthropic CEO Dario Amodei subsequently urged an industrywide slowdown in advanced model development efforts globally.
OpenAI CEO Sam Altman and Tesla and SpaceX CEO Elon Musk also supported calls for slower development, according to the report. The companies’ simultaneous product launches, however, indicate that commercial development continues while executives, researchers and policymakers consider how safety concerns should affect the industry’s future direction and operating standards.
Cost pressure is a central factor behind the new offerings. Corporate customers are attempting to control rising AI expenses, while OpenAI and Anthropic face competitors providing cheaper open-weight systems. These include Chinese technology companies Alibaba, Moonshot AI and DeepSeek. The announcements reflect competition over pricing, efficiency, performance and safety priorities.