Model Releases

OpenAI releases medical evaluation results for GPT-5.6 series

Original title:"physicians found fewer flaws in GPT-5.6 responses than physician-written responses."

AI summary

OpenAI released evaluation results for the GPT-5.6 series in the medical field. The smallest variant, GPT-5.6 Luna, surpasses GPT-5.5 at its highest reasoning effort even at the lowest reasoning effort, at 25 times lower cost; the largest variant, GPT-5.6 Sol, sets a new bar. In diverse tasks covering both patient-facing and clinical settings, specialist physicians were asked to write responses with unlimited time and internet access, which were then blindly evaluated by other physicians. The evaluation covered five dimensions—accuracy, communication, completeness, instruction following, and helpfulness for health decisions—with 20,000 ratings in total. Results showed that all GPT-5.6 models performed significantly better than physicians, and physicians found fewer flaws in GPT-5.6 responses than in their own.

Key Takeaways

  • OpenAI released evaluation results for the GPT-5.
  • 6 series in the medical field.
  • The smallest variant,GPT-5.
  • 6 Luna,surpasses GPT-5.
  • 5 at its highest reasoning effort even at the lowest reasoning effort,at 25 times lower cost

Quick Facts

Published
Updated
Source
X:Sam Altman (@sama)
Category
Model Releases
AI score
68
Read the original Read the full article at the original source
Why it matters

GPT-5.6 在医疗任务上被医生评价比真人医生缺陷更少, 这是AI辅助诊断的分水岭, 产品人和医疗从业者值得看具体数据。

Related entities

Source & copyright

This article is based on curated titles and AI summaries from the AIHOT public API, with attribution and links back. Original copyright remains with the original publishers and authors; this site does not mirror full texts.

Original article · AIHOT entry

How to Cite

When citing this article, keep the title, date, and permalink, and credit Orange News as the source.

Orange News: OpenAI releases medical evaluation results for GPT-5.6 series. July 12, 2026. https://new.nat6.net/en/news/cmrglvcki024uih2edvsk7pef/

View all →