As the LLM competition heats up, we're seeing AI tech companies burn through version numbers like they were made of wood. Any signs of slacking or lagging behind may permanently set the company back relative to its peers, so everyone is doing their best to release new models regularly.

OpenAI is no exception, and now, the company has just released GPT-5.5. And if you have a strange feeling that GPT-5.4 surely can't be that old, well, you're right; but that hasn't stopped the company from making a new one anyway.

OpenAI has released GPT-5.5, and you can use it right now

Well, that was fast

Over on X, OpenAI has revealed that GPT-5.5 is now available for general use. It follows hot on the heels of GPT-5.4, which was only just released in early March. It's a similar pattern we're seeing with Anthropic, where a model will barely see its two-month anniversary before a successor steps in.

So, just how much can OpenAI do in under two months? Well, in a reply to the original post, the company shared some benchmarks versus its older model, and you can see the results in the image above. The biggest jump was with Thinking with Terminal-Bench 2.0, with GPT-5.5 outperforming the previous model by over 7%. OpenAI also adds competitor benchmarks to the chart, indicating that the company believes ChatGPT is either comparable to or beating the competition.

OpenAI says you'll more likely see gains in "agentic coding, computer use, knowledge work, and early scientific research," which the company says are all areas where the AI has to progress step by step to achieve something. If you want to see whether the AI model lives up to its claims, you can give it a try today if you're a Plus, Pro, Business, or Enterprise subscriber. It'll appear in both ChatGPT and Codex.