Model Release

OpenAI Launches GPT-5.5 Instant for Enhanced Health Intelligence

OpenAI's GPT-5.5 Instant improves health-related responses, matching frontier models' performance in critical evaluations, with a 71% drop in factuality issues over two months.

Image: OpenAI

OpenAI has introduced GPT-5.5 Instant, an enhanced version of its ChatGPT model, designed to improve health intelligence and support users with medical inquiries. Every week, more than 230 million people use ChatGPT for health and wellness advice, ranging from interpreting lab results to preparing for medical appointments. GPT-5.5 Instant represents a significant step forward in health-related interactions, offering better recognition of urgent care needs and clearer explanations of complex information. This model is now available to all free users in ChatGPT, allowing a broader audience to benefit from these improvements. Source: openai

The progress in health intelligence reflects both advancements in model capabilities and the involvement of a global network of physicians. These professionals help define what constitutes a 'good' response in real-world health scenarios by reviewing example model responses, describing ideal behavior, and identifying failure modes. This collaboration enables OpenAI to measure and improve the model's performance over time. Source: openai

To assess health-related performance, OpenAI uses evaluations like HealthBench and HealthBench Professional, which incorporate realistic health conversations and physician-written rubrics. GPT-5.5 Instant now performs at a level comparable to its frontier models on these evaluations, showing a substantial improvement from GPT-5.3 Instant. Physicians also rated GPT-5.5 Instant responses as having fewer failure modes than those from older models and physicians, particularly in areas like tailoring to local healthcare contexts and identifying red flags. Source: openai

Source: openai

Key points

GPT-5.5 Instant matches frontier models' performance on health evaluations, including HealthBench Professional.
Physicians rated GPT-5.5 Instant responses as having fewer failure modes than those from older models and physicians.
The rate of health responses with at least one flagged factuality issue has fallen by 71% in the last two months.
GPT-5.5 Instant is available to all free users in ChatGPT, allowing broader access to improved health intelligence.
HealthBench and HealthBench Professional evaluations use realistic health conversations and physician-written rubrics to assess model responses.
GPT-5.5 Instant shows substantial improvement from GPT-5.3 Instant on health evaluations.

Source: OpenAI Read the original →

WRITTEN BY

Alex Lindgren

LLMs & Frontier Models

Alex covers the large language models and their impact on society.