OpenAI launches its new family of models with GPT-5.6

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

GPT-5.6 is available in three variants: Sol (considered its workhorse), Terra (a more intermediate option), and Luna (the budget-friendly option). These models expand what users can do in different areas – with the company promising powerful capabilities in entrepreneurial work, coding and even scientific research.

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

Most notably, the company is calling 5.6 its “strongest cybersecurity model yet, achieving groundbreaking performance with significantly fewer tokens.”

Indeed, there has been a lot of fuss about the model’s cyber capabilities, as the Trump administration previously tried to limit its rollout, ostensibly out of fear of the model being 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 has also released a new tool called ChatGPT Workwhich – as it sounds – is designed as a workplace companion for business teams, running across desktop, web and mobile, that can assist with daily administrative tasks such as drafting documents, spreadsheets and presentations.

OpenAI’s newly announced family of models follows similar releases this week from competitors SpaceXAI and Meta.

However, GPT-5.6 and its associated marketing seem primarily intended to target OpenAI’s main opponent, Anthropic. Anthropic has managed to make itself the likeable underdog of the AI ​​race, focusing heavily on enterprise customers and gaining an increasing share of support as a result.

Not to be outdone, OpenAI calls the Artificial Analysis Encryption Agent Indexa notable benchmarking metric, to claim that the latest model family outpaces Anthropic’s models at every turn.

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 “ranks a new state of play 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.”

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

The company says 5.6 is now available via ChatGPT, Codex and the OpenAI API. The 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.