OpenAI launches GPT-5.5, bringing the company one step closer to an AI “super application.”

OpenAI on Thursday released GPT-5.5, its latest AI model, which the company calls “the smartest and most user-friendly model” to date. The algorithm comes with increased capabilities in several areas, with OpenAI co-founder and president Greg Brockman claiming that it also brings the company one step closer to creating a “super app” for OpenAI.

On a call with reporters, Brockman said the new model represents major progress “toward more efficient and intuitive computing.”

“This model is a real step forward toward the kind of computing we expect in the future, but it’s one step, and we expect to see a lot more of it in the future,” Brockman said. “It’s faster and clearer thinking for fewer codes than something like 5.4. This means there is more frontier AI available to businesses and consumers, which is part of our goal.”

Brockmann also said the model was an additional step toward creating a “super app” — a Swiss Army multi-purpose program — which Brockmann and co-founder Sam Altman had previously discussed launching. The founders envision combining ChatGPT browser, Codex, and AI into one unified service that can help enterprise customers. Notably, the “super app” concept is also a hot topic with Altman’s competitor (and former OpenAI colleague) Elon Musk, who has said he wants to turn X into his own so-called super app.

OpenAI released its latest model just last month, with a previous release in December, and before that in November. The company has continued to produce new models at a rapid pace, a trend that company employees said is expected to continue for the foreseeable future. “We see very significant improvements in the short term, and very significant improvements in the medium term,” said Jacob Paczucki, chief scientist at OpenAI. “Actually, I would say, I think the last couple of years have been surprisingly slow.”

According to OpenAI, GPT-5.5 is designed to be useful across a wide range of categories, including core enterprise areas like agent cryptography and cognitive work, but also in more experimental AI applications like mathematics and scientific research. The company also released data on Thursday showing the model’s outperformance across a range of benchmarks. Compared to its predecessors, and models from competitors Google and Anthropic (such as the Gemini 3.1 Pro and Claude Opus 4.5), the 5.5 scores consistently higher, according to OpenAI.

The rivalry between OpenAI and Anthropic is not far from being discussed, and GPT-5.5 has presented another opportunity for the two companies to compare themselves against each other. A reporter asked during the press conference if it was GPT 5.5. It will have capabilities similar to Mythos, the cybersecurity tool Anthropic recently announced. (Mythos has faced controversy in recent days over a report of unauthorized access to the software.) Mia Glaese, a technical staff member at OpenAI, said GPT-5.5 will have a major impact on the company’s approach to deploying its models toward digital defense. “We have a strong, long-term strategy for our approach to cyber, and have continually refined our approach to rolling out models securely,” Glaze said.

Mark Chen, chief research officer at OpenAI, said GPT-5.5 was better at navigating computer work than its predecessors, and also said the model “demonstrates meaningful gains in scientific and technical research workflows,” noting that the company feels it can truly “help expert scientists make progress.” Chen also said it could help in drug discovery, an area that has shown increasing interest in the industry over the past few years.

GPT 5.5 is widely available starting Thursday, according to OpenAI. The company says the model is being rolled out to Plus, Pro, Business, and Enterprise users in ChatGPT, while Pro 5.5 is headed to Pro, Business, and Enterprise users.