OpenAI has replaced the engine inside ChatGPT. Starting today, GPT-5.5 Instant becomes the default model for hundreds of millions of users worldwide, succeeding GPT-5.3 Instant.
This isn't a flashy launch. But the upgrade cuts one of ChatGPT's most persistent flaws-hallucinations-by more than half. In OpenAI's internal tests, the new model produced 52.5% fewer fabricated claims on high-stakes prompts in medicine, law, and finance. On real user conversations previously flagged for factual errors, inaccurate claims dropped by 37.3%.
On HealthBench, which tests AI responses to real medical questions, GPT-5.5 Instant scored 51.4 out of 100, up from 49.6. On the professional clinical version, it jumped from 32.9 to 38.4 points. Health questions remain among the most common queries users ask ChatGPT.
GPT-5.5 Instant also improves personalization by pulling context from past chats, saved files, and connected Gmail accounts. Users can now see exactly which context was used and delete or correct it.
Notably, GPT-5.5 Instant is the first Instant-tier model classified as "High Capability" in both cybersecurity and biological domains-meaning it requires automated safeguards previously reserved for more powerful variants. Paid subscribers who prefer the old version have three months before GPT-5.3 Instant is retired.