OpenAI GPT-5.6 Launch: AI Spend & Efficiency Focus

AI spend is the big topic at this year's Allen & Co. conference in Sun Valley. OpenAI CEO Sam Altman says cost and return on investment are the dominant questions among executives. This is the first year AI spending has been a "very big topic" at the summit. Everyone wants to know how to reduce spend or increase value. OpenAI sees enterprises prioritizing efficiency and demonstrable returns from their AI investments. This cost pressure directly influenced the new GPT-5.6 model family, which OpenAI released publicly. It includes three tiers: Sol, the flagship; Terra, a balanced option; and Luna, the cost-efficient tier. A limited preview began for partners on June 26th, with general availability on July 9th. The names are deliberate. The number identifies the generation, while Sol, Terra, and Luna identify capability tiers that can advance independently. OpenAI also released ChatGPT Work, an agent designed to handle entire jobs. Altman states GPT-5.6 Sol is "not only the best model in the world for most people" and is 54% more token efficient for agentic coding tasks. This public launch follows a period of restricted access for trusted partners. Every enterprise is now focused on the value they get for their AI spend. This shift is evident across the industry, with executives like Coinbase CEO Brian Armstrong exploring cheaper models to optimize costs.