OpenAI announced on Thursday that it is reducing the prices of its two newest artificial intelligence models, GPT-5.6Terra and GPT-5.6Luna, just roughly three weeks after their public launch.

Enterprise clients are becoming increasingly sensitive to costs and are reluctant to deploy expensive models unless they can clearly see a return on investment. OpenAI is under pressure to meet the needs of such customers. Simultaneously, the company must defend against competition from tech giants like Google and Microsoft, which have been aggressively promoting high-value, cost-effective models.

On Thursday, the company stated that the price of Terra has been cut by 20%. Its input token fee has dropped to $2 per million tokens, while the output token fee is set at $12 per million tokens. Luna has seen an 80% price reduction, with input token fees falling to $0.2 per million tokens and output token fees to $1.2 per million tokens. Pricing for Sol remains unchanged.

In its announcement, OpenAI said: "Our core strategy remains to simultaneously improve model capabilities and operational efficiency, striving for each generation of intelligent models to accomplish more tasks at a lower cost."

The launch of ChatGPT in 2022 ignited a surge in artificial intelligence, driving businesses across the United States to rush to adopt the technology and encourage widespread employee use, ushering in what is known as the "maximize token consumption" phase. At that time, companies encouraged employees to use AI as much as possible without worrying about costs. However, as AI-related expenses have skyrocketed, with some companies facing costs reaching billions of dollars, many organizations are now actively working to control spending.