The AI arena is free today

Open Superagent

HealthBench Professional

Paper

Progress Over Time

Interactive timeline showing model performance evolution on HealthBench Professional

State-of-the-art frontier
Open
Proprietary

HealthBench Professional Leaderboard

22 models
ContextCostLicense
1—1.0M$2.00 / $10.00
2
Anthropic
Anthropic
—1.0M$8.00 / $40.00
3—1.0M$10.00 / $50.00
4—1.0M$4.00 / $20.00
5—1.0M$0.10 / $0.50
6—1.1M$2.00 / $10.00
7—1.1M$10.00 / $50.00
8—1.1M$1.00 / $5.00
9———
9———
11—1.0M$4.00 / $20.00
11———
11—1.1M$5.00 / $30.00
14
Alibaba Cloud / Qwen Team
Alibaba Cloud / Qwen Team
2.4T984K$1.50 / $4.50
15
Anthropic
Anthropic
—1.0M$5.00 / $25.00
16—1.0M$2.00 / $10.00
17—1.1M$2.00 / $12.00
18—500K$2.00 / $6.00
19—1.0M$5.00 / $25.00
20—1.1M$0.20 / $1.20
21—400K$5.00 / $30.00
221.0T——
Notice missing or incorrect data?

Sub-benchmarks

About this benchmark

What is HealthBench Professional?

HealthBench Professional evaluates model capability and safety for clinician use cases using real clinician-style chats and physician-authored grading rubrics.

HealthBench Professional is a text benchmark evaluating models on healthcare tasks. LLM Stats tracks 22 models on this benchmark, scored on a 0–1 scale. The current average is 0.6, with the leader at 0.7.

Compare leaders on the best AI for healthcare leaderboards.

Current leaders

Claude Sonnet 5.5 from Anthropic currently leads the HealthBench Professional leaderboard with a score of 0.692 across 22 evaluated AI models.

1Claude Sonnet 5.5Anthropic69.2%
2Claude Fable 5.2Anthropic67.7%
3Claude Fable 5Anthropic66.0%

FAQ

Common questions about the HealthBench Professional benchmark and leaderboard.

What is the HealthBench Professional benchmark?

HealthBench Professional evaluates model capability and safety for clinician use cases using real clinician-style chats and physician-authored grading rubrics.

What is the HealthBench Professional leaderboard?

The HealthBench Professional leaderboard ranks 22 AI models based on their performance on this benchmark. Currently, Claude Sonnet 5.5 by Anthropic leads with a score of 0.692. The average score across all models is 0.594.

What is the highest HealthBench Professional score?

The highest HealthBench Professional score is 0.692, achieved by Claude Sonnet 5.5 from Anthropic.

How many models are evaluated on HealthBench Professional?

22 models have been evaluated on the HealthBench Professional benchmark, with 0 verified results and 22 self-reported results.

Where can I find the HealthBench Professional paper?

The HealthBench Professional paper is available at https://cdn.openai.com/dd128428-0184-4e25-b155-3a7686c7d744/HealthBench-Professional.pdf. The paper details the methodology, dataset construction, and evaluation criteria.

What categories does HealthBench Professional cover?

HealthBench Professional is categorized under healthcare. The benchmark evaluates text models.

What's the difference between HealthBench Professional and HealthBench?

HealthBench Professional is a variant of HealthBench. See the HealthBench leaderboard for the broader benchmark and per-model comparison.

Are there variants of HealthBench Professional?

Yes. HealthBench Professional has 1 related variant: HealthBench Professional (raw).

Which model offers the best value on HealthBench Professional?

Among models scoring within 10% of the leader, Claude Haiku 5.5 from Anthropic is the cheapest, at $0.10 per million input tokens with a score of 0.648.

How recent are the HealthBench Professional leaderboard results?

The HealthBench Professional leaderboard was last updated in November 2026 and currently includes 22 evaluated models.