OpenAI launches its new set of models with GPT-5.6

OpenAI unveiled its latest batch of models on Thursday, introducing a new set of software heavyweights in an increasingly crowded field of AI offerings.

GPT-5.6 comes in three variants: Sol (its workhorse), Terra (its more intermediate option), and Luna (its budget-friendly option). These models expand what users can do across a variety of fields – with the company promising powerful capabilities in enterprise work, programming, and even scientific research.

CEO Sam Altman has promised that his company’s latest models are an order of magnitude more efficient and cost-effective than previous versions, recently telling CNBC that Sol is 54% more code-efficient when it comes to AI coding tasks.

Notably, the company describes 5.6 as “the most powerful cybersecurity model to date, achieving frontier performance using far fewer codes.”

In fact, there has been so much buzz around the model’s cyber capabilities that the Trump administration has previously sought to restrict its rollout, ostensibly out of concerns about how the model could be misused. GPT-5.6 supports defensive activities, including threat modeling, code review and debugging, and blue-collaring (simulating an attack on your own systems to find vulnerabilities before real hackers do).

OpenAI has also released a new tool called ChatGPT Work, which – it turns out – is designed to be a workplace companion for enterprise teams, running on desktop, web and mobile, and which can help with everyday clerical tasks, such as drafting documents, spreadsheets and presentations.

The newly announced OpenAI model suite comes on the heels of similar releases this week from competitors SpaceXAI and Meta.

However, GPT-5.6 and its accompanying marketing appear designed to target OpenAI’s primary opponent, Anthropic. Anthropic has managed to position itself as the darling underdog in the AI ​​race, steadily focusing on enterprise customers and winning an increasing share of support as a result.

Not to be outdone, OpenAI cites the Synthetic Analysis Coding Factor Index, a prominent benchmark, to claim that its latest suite of models outperforms Anthropic’s models at every turn.

OpenAI calls Sol “the best programming model yet,” and has pointedly compared it to Anthropic’s recently released (and much-hyped) Fable. Using a cryptographic proxy indicator, OpenAI claims that Sol “sets a new state of progress at 80, 2.8 points above Fable 5, while using less than half the output codes, taking less than half the time, and costing about a third less.”

“This advantage extends to the whole family: Terra performs slightly above Fable 5, while Luna outperforms Opus 4.8,” he adds.

The company says that version 5.6 is now available via ChatGPT, Codex, and OpenAI API. Availability per million tokens is priced as follows: Sol is $5 input/$30 output, Terra is $2.50 input/$15 output, and Luna is $1 input/$6 output.