OpenAI Introduces GPT-5.5: A Leap Forward in Multi-Step Reasoning
OpenAI has recently unveiled GPT-5.5, hailed as the company’s “smartest and most intuitive model yet.” This announcement comes merely six weeks after the release of GPT-5.4, reflecting the rapid pace of AI development.
The new model shines in executing complex, multi-step tasks such as writing and debugging code, conducting online research, analyzing data, and operating software autonomously. Unlike its predecessors that required meticulous step-by-step guidance, GPT-5.5 can independently plan its approach, use tools, check its own work, and navigate through ambiguity until tasks are successfully completed.
Significant improvements have been observed in GPT-5.5 during benchmark testing. It leads the competition in areas like terminal tasks (scoring 82.7% on Terminal-Bench 2.0), expert software engineering, and cybersecurity simulations. Notably, the model outperforms Anthropic’s Claude Opus 4.7 and Google’s Gemini 3.1 Pro across most categories, all the while maintaining similar latency to previous versions but utilizing fewer tokens.
As of now, GPT-5.5 is available to ChatGPT Plus, Pro, Business, and Enterprise users. API access will be available from April 24. OpenAI also announced that ChatGPT now boasts over 900 million weekly active users, more than 50 million subscribers, and 9 million paying business users. The company emphasized that GPT-5.5 includes its most robust safety measures to date, with enhanced safeguards for cybersecurity and biological risks.
Source: https://openai.com/index/introducing-gpt-5-5/