OpenAI says its new GPT-5.5 Instant model significantly reduces hallucinations while improving context awareness and everyday task performance

Prompt hallucinations remain an ongoing challenge for AI models, but OpenAI says its latest GPT-5.5 Instant model shows significant improvements in maintaining factual accuracy. According to the company’s internal evaluations, GPT-5.5 Instant produces 52.5% fewer hallucinated claims compared to GPT-5.3 Instant in high-stakes domains such as medicine, law, and finance. Additionally, inaccurate claims have been reduced by 37.3% in challenging conversations previously flagged by users for factual errors.

Beyond accuracy, OpenAI states that GPT-5.5 Instant is more capable in everyday tasks, including analyzing image uploads and determining when to use the web for answers. The model is also more context-aware, even when referencing previous conversations.

STORY CONTINUES BELOW THIS AD

Instant now makes better use of context from past chats, files, and even Gmail (if connected), delivering more personalized responses while keeping users in control. It can intelligently determine when additional personalization will improve an answer and retrieves relevant context from previous conversations more quickly, reducing the need to repeat information. This is particularly useful for tailored recommendations, planning, and seamlessly continuing ongoing work.

OpenAI has further introduced a new “Memory Source” feature across ChatGPT models, allowing users to see what context is used to generate personalized responses. Users can also correct or delete this information as needed.

The GPT-5.5 Instant model will be rolled out to the public on Tuesday, while GPT-5.3 Instant will remain available for three months before being retired. These enhanced personalization features will first roll out to Plus and Pro users on the web, followed by mobile applications. OpenAI has also indicated plans to extend access to free, Go, Business, and Enterprise users in the near future.

First Published:
May 06, 2026, 00:00 IST

HomeTechOpenAI claims ChatGPT’s latest default model cuts hallucinations significantlyEnd of Article