Skip to content
Trending storyDeveloping

Artificial Analysis launches HealthBench and HealthBench Professional leaderboards

1 article1 sourcesince Oct 19Last article 1h ago ·

Overview

AISummary of 1 article

Artificial Analysis launches leaderboards for HealthBench and HealthBench Professional, two OpenAI benchmarks that grade AI model replies to health conversations against physician-written rubrics.

On HealthBench Professional, GPT-6 Astra leads with 66.0% on the length-adjusted score, and GPT-6 Luna reaches 61.9% at under 2% of its cost, according to Artificial Analysis. On HealthBench, Claude Sonnet 5.5 leads with 62.1%, ahead of DeepSeek V4.1 Flash at 60.9%.

Written by AI from the articles below · updated Oct 19, 7:17 PM ET

Check the sources:

Latest developmentOct 19, 7:03 PM ET
Artificial Analysis launches HealthBench and HealthBench Professional leaderboards

Article timeline

The articles in this story. Times are ET.

Oct 19
  1. Artificial AnalysisOfficialPick
    Artificial Analysis launches HealthBench and HealthBench Professional leaderboards

    AIArtificial Analysis launches leaderboards for HealthBench and HealthBench Professional, OpenAI benchmarks that grade model replies to health conversations against physician-written rubrics. HealthBench Professional leader GPT-6 Astra scores 66.0% on the length-adjusted score, and GPT-6 Luna reaches 61.9% at under 2% of its cost. On HealthBench, Claude Sonnet 5.5 leads with 62.1%, ahead of DeepSeek V4.1 Flash at 60.9%.

    Image from @ArtificialAnlys's post

Heat Trend

Not enough continuous observations to show a trend yet.