OpenAI Launches GPT-5.5, Its ‘Smartest Model Yet’ for Coding, Research & Autonomous Work

OpenAI has introduced GPT-5.5, a major upgrade to its flagship AI model, positioning it as its most capable system yet for complex, real-world tasks such as coding, research, and enterprise workflows.

The release marks a continued push toward more “agentic” AI systems that can plan, execute, and complete tasks with minimal human input.

According to the company, GPT-5.5 is designed to handle “complex goals, use tools, check its work, and carry more tasks through to completion,” reflecting a shift from simple question-answering toward systems that can operate more like autonomous digital workers.

The model shows significant improvements in professional use cases, particularly in software development, data analysis, and document-heavy workflows. OpenAI said GPT-5.5 performs strongly in tasks such as coding, research, and information synthesis, delivering faster and more concise responses for complex problems.

"It is smart and fast; per-token speed matches 5.4 and it uses significantly fewer tokens per task. In my experience, it "gets what to do". Rolling out today in ChatGPT and Codex. We are working with API customers on security and safeguards and plan to launch in API very soon," Sam Altman, OpenAI CEO, said.

OpenAI also highlighted GPT-5.5’s ability to persist through multi-step workflows, making it more effective in scenarios that require continuous reasoning, such as scientific research.

“It’s incredibly energising to use OpenAI’s new GPT-5.5 model in our harness, have it reason over massive biochemical datasets to predict human drug outcomes, and then see it deliver significant accuracy gains on our hardest drug discovery evals. If OpenAI keeps cooking like this, the foundations of drug discovery will change by the end of the year,” Brandon White, Axiom Bio Co-Founder & CEO, said.

In internal and early testing, the model demonstrated strong performance across benchmarks tied to real-world knowledge work, including operating software environments and handling customer-service workflows.

The company said early users are already leveraging the model as a collaborative partner rather than a one-off tool—using it to critique research, analyze datasets, and iterate on complex problems over multiple cycles. In one example, researchers used GPT-5.5 to generate insights from large-scale genetic data, significantly reducing the time required for analysis.

“GPT-5.5 is noticeably smarter and more persistent than GPT-5.4, with stronger coding performance and more reliable tool use. It stays on task for significantly longer without stopping early, which matters most for the complex, long-running work our users delegate to Cursor,” Michael Truell, Cursor Co-founder & CEO, said.

GPT-5.5 also introduces improved efficiency and reasoning capabilities, enabling it to better understand user intent, require less guidance, and complete tasks with fewer errors.

"We believe in iterative deployment; although GPT-5.5 is already a smart model, we expect rapid improvements. Iterative deployment is a big part of our safety strategy; we believe the world will be best equipped to win at the team sport of AI resilience this way," Altman added.