Major US artificial intelligence companies including OpenAI and Anthropic have announced significant price reductions for their mid-range models in response to mounting competitive pressure from Chinese AI developers. The pricing adjustments come as corporations increasingly seek cost-effective alternatives to manage rising AI expenses, with some major firms reportedly adopting models from Chinese competitors such as DeepSeek and Moonshot to control costs.
OpenAI reduced prices for its GPT-5.6 Luna model by 80 percent, lowering input token costs from $1 to $0.20 per million tokens and output token costs from $6 to $1.20 per million. Anthropic introduced Claude Opus 5, priced at half the cost of its more advanced Fable 5 model, while also canceling a scheduled price increase for its Sonnet 5 offering. According to token pricing indices, these moves have contributed to nearly a 25 percent overall reduction in model costs since mid-July.
The pricing war reflects a strategic shift for US AI laboratories that have traditionally competed on performance capabilities rather than cost. Chinese developers have released increasingly capable open-source models that can be downloaded and customized by users, creating new competitive dynamics. Major corporations including DoorDash and Airbnb have adopted Chinese-made models to manage AI-related expenditures, raising concerns within the US technology sector about market share erosion.
These developments occur as OpenAI and Anthropic prepare for initial public offerings valued at trillion dollars, intensifying pressure to demonstrate profitability and return on massive infrastructure investments. Enterprise customers are also shifting from flat-rate subscription models to usage-based billing arrangements, creating incentives for organizations to reduce computational consumption. Industry analysts note that while headline token prices have decreased substantially for mid-tier offerings, actual per-task costs depend on model efficiency and computational effort settings used.
Article Attribution | Read More at Article Source
Article summary produced by Claude AI