OpenAI launches its new family of models with GPT-5.6

OpenAI unveiled its newest family of models on Thursday, introducing GPT-5.6 with three variants: Sol (the workhorse), Terra (intermediate), and Luna (budget-friendly). The company says these models expand capabilities across enterprise work, coding, and scientific research.

Efficiency and cybersecurity

CEO Sam Altman has stated that the new models are orders of magnitude more efficient and cost-effective than previous versions. He told CNBC that Sol is 54% more token efficient for AI coding tasks. OpenAI calls GPT-5.6 its “strongest cybersecurity model yet,” achieving frontier performance with significantly fewer tokens. The Trump administration had previously sought to restrict its rollout due to misuse fears. The models support defensive activities such as threat modeling, code review and patching, and blue teaming.

ChatGPT Work and competition

OpenAI also launched ChatGPT Work, a workplace companion for enterprise teams running on desktop, web, and mobile, assisting with clerical tasks like drafting documents, spreadsheets, and presentations. The new model family follows similar releases this week from SpaceXAI and Meta. However, GPT-5.6 seems aimed at primary rival Anthropic. Citing the Artificial Analysis Coding Agent Index, OpenAI claims its Sol model outshines Anthropic’s Fable 5 by 2.8 points, while using less than half the output tokens, taking less time, and costing about one-third less. Terra performs just above Fable 5, and Luna outperforms Opus 4.8.

Availability and pricing

GPT-5.6 is available on ChatGPT, Codex, and the OpenAI API. Pricing per million tokens: Sol – $5 input / $30 output; Terra – $2.50 input / $15 output; Luna – $1 input / $6 output.