Tech
giant OpenAI has introduced GPT-5.5 Instant, its upgraded and default model in ChatGPT. The roll out of this new AI model began today for all the users globally. With this update, the Sam Altman-led firm wants to make its chatbot responses smarter, more accurate and easier to understand. The company highlights that this new model will improve interactions by reducing errors and now uses less emojis. Here is how it compares to its previous iteration, GPT-5.3.
How GPT-5.5 Instant Works
As per the company,
GPT-5.5 Instant can show significant improvements in accuracy, especially in areas like law, finance and medicine. This model is claimed to offer reduced hallucinations by over 50 per cent compared to GPT-5.3 Instant. It can also reduce incorrect answers in conversations that were previously flagged as errors most of the time. This model is better at finding its own mistakes and correcting them. For instance, it can revisit eros in problem-solving tasks and arrive at the accurate solution instead of stopping in the middle. Speaking of
personalisation, the OpenAI has noted that the upgraded model now offers clearer, shorter and more relevant responses without losing crucial details. The chatbot also uses past context to offer personalisation and helpful suggestions. These may involve using previously uploaded files and even connected Gmail data to offer a more tailored response. These features will be rolled out to Plus and Pro users on the web and soon be accessible on mobiles and other plans. It is notably that the San Francisco-based company asserted how the tone of this AI model has become more conversational and natural. Moreover, it will now avoid unnecessary follow-up questions and reduce excessive emojis in its formatting.
Who Can Use GPT-5.5 Instant
The new and upgraded
ChatGPT model is gradually becoming the default for all ChatGPT users. The earlier version, GPT-5.3 Instant, will still be available for paid users for the next three months before it gets phased out.