SAN FRANCISCO, May 5 — OpenAI on Monday released GPT-5.5 Instant, a new model that replaces GPT-5.3 Instant as the default in ChatGPT, the company said in a blog post.
The lab said internal evaluations show GPT-5.5 Instant producing 52.5% fewer hallucinated claims than GPT-5.3 Instant on high-stakes prompts in areas including medicine, law, and finance. OpenAI described the update as smarter, more accurate, and offering improved personalization controls.
The model is rolling out to all ChatGPT users beginning Monday and replacing GPT-5.3 Instant as the default, the company said.
In a separate set of announcements on the same day, OpenAI introduced new realtime voice intelligence models in its API capable of reasoning, translation, and transcription within a single inference path; opened a preview of a personal-finance experience inside ChatGPT for U.S. Pro subscribers; and added Codex to the ChatGPT mobile application in preview, with extended remote-SSH, hooks, access-token, and HIPAA support for enterprise teams.
OpenAI did not publish third-party benchmark scores at launch and did not disclose the internal evaluation methodology behind the hallucination-reduction figure beyond stating the prompt categories covered.
The release lands two weeks before Alphabet’s Google announced Gemini 3.5 Flash, Gemini Omni, and a new general-purpose Gemini Spark agent at its annual developer conference on May 19.
OpenAI did not provide further guidance on the timing of a successor to its Pro-tier reasoning models.