OpenAI unveiled its new family of models on Thursday, introducing a new set of heavyweight programs into an increasingly crowded field of AI offerings.

GPT-5.6 comes in three variants: Sol (considered your workhorse), Terra (a more middle-of-the-road option), and Luna (your budget option). These models expand what users can do in a variety of fields, and the company promises powerful capabilities in business work, coding, and even scientific research.

CEO Sam Altman has promised that his company’s newest models are much more efficient and cost-effective than previous versions, and recently told CNBC that Sol is 54% more symbolically efficient when it comes to AI coding tasks.

Notably, the company calls 5.6 its “most robust cybersecurity model yet, achieving cutting-edge performance with significantly fewer tokens.”

In fact, there has been a lot of fuss about the model’s cyber capabilities, as the Trump administration previously attempted to restrict its implementation, apparently due to fears of how the model could be misused. GPT-5.6 supports defensive activities, including threat modeling, code review and patching, and blue teaming (simulating an attack on your own systems to find weaknesses before real hackers do).

OpenAI also launched a new tool called ChatGPT Work, which, just as it sounds, is designed as a work companion for enterprise teams, running on desktop, web, and mobile devices, and can help with daily administrative tasks such as drafting documents, spreadsheets, and presentations.

OpenAI’s recently announced model family follows on the heels of similar launches this week from competitors SpaceXAI and Meta.

However, GPT-5.6 and its accompanying marketing seem more designed to target OpenAI’s main opponent, Anthropic. Anthropic has managed to become the sympathetic loser in the AI ​​race, focusing squarely on enterprise customers and winning a growing share of support as a result.

Not to be outdone, OpenAI cites the Artificial Analysis Coding Agent Index, a notable benchmarking metric, to claim that its latest family of models outshines Anthropic’s models every time.

OpenAI calls Sol its “best coding model yet” and has explicitly compared it to Anthropic’s recently released (and much-hyped) Fable. Using the Coding Agent Index, OpenAI claims that Sol “sets a new state of the art at 80, 2.8 points above Fable 5, while using less than half the output tokens, taking less than half the time, and costing about a third less.”

He adds: “That advantage extends to the entire family: Terra performs just above Fable 5, while Luna outperforms Opus 4.8.”

The company says that 5.6 is now available on ChatGPT, Codex, and the OpenAI API. Availability per million tokens is priced as follows: Sol costs $5 in/$30 out, Terra costs $2.50 in/$15 out, and Luna costs $1 in/$6 out.