HealthBench Professional Leaderboard
Length-adjusted score
The score the board ranks on: rubric points earned, adjusted for answer length as OpenAI reports it. 43 models, 525-task set, compiled from published documents.
- OpenAI
- Anthropic
- Meta
- SpaceX AI
- Moonshot AI
- Zhipu
- DeepSeek
- Microsoft
GPT-6 Astra (Anthropic run)
Claude Sonnet 5.5
Claude Fable 5
Claude Opus 5.5
GPT-6 Luna (Anthropic run)
Claude Haiku 5.5
GPT-6 Astra
GPT-6.1 Sol
Claude Fable 5 (September card)
GPT-6.1 Terra
Muse Wonder Full
Claude Fable 5.1
GPT-6 Luna
GPT-6 Sol
GPT-5.6 Sol
Muse Wonder
Claude Opus 5
Muse Spark 1.1
Claude Sonnet 5
GPT-5.6 Terra
Claude Opus 4.8 (Opus 4.8 grader)
Grok 4.7
Claude Opus 4.8
GPT-5.6 Luna
Gemini 4 Argon
Muse Spark
GPT-5.6 Sol (August)
Kimi K3
GLM 5.3
Claude Opus 4.7
GPT-5.5
Gemini 3.8 Flash
- DDeepSeek V4.1 Flash
Grok 4.6
GPT-5.4
GPT-5
GPT-5.2
Claude Sonnet 4.6
GPT-5.6 Luna (August)
GPT-5.1
GPT-5.5 Instant
MAI-Thinking-1
Claude Haiku 4.5
physician-written responses 0.437
Score before the length adjustment
Rubric points earned with no adjustment for answer length, for the 30 of 43 models whose source prints it. Longer answers earn more here, so the order differs from the board.
- OpenAI
- Anthropic
- Moonshot AI
- Zhipu
- DeepSeek
Claude Sonnet 5.5
Claude Opus 5.5
Claude Fable 5.1
GPT-6 Astra (Anthropic run)
Claude Opus 5
Claude Haiku 5.5
Claude Fable 5
Claude Fable 5 (September card)
GPT-6 Astra
GPT-6.1 Sol
GPT-6.1 Terra
Gemini 4 Argon
GPT-5.6 Sol
Claude Sonnet 5
GPT-5.6 Terra
GPT-6 Luna
GPT-5.6 Luna
GPT-6 Sol
Gemini 3.8 Flash
GLM 5.3
GPT-5.5
Kimi K3
GPT-5.6 Sol (August)
- DDeepSeek V4.1 Flash
GPT-5.4
GPT-5
GPT-5.2
GPT-5.1
GPT-5.6 Luna (August)
GPT-5.5 Instant
Full ranking
Click a column to sort. Each model links to its full result; the line under the name says which document the score was read from. List API prices per 1M tokens as of November 2, 2026. GPT-5.6 models charge higher rates above 272K input tokens. MAI-Thinking-1 is in public preview on Microsoft Foundry without final list pricing.
| License | ||||||||
|---|---|---|---|---|---|---|---|---|
| 1 | 0.703 | 0.740 | 1.05M | $10.00 | $50.00 | 2026-09-03 | proprietary | |
| 2 | 0.692 | 0.771 | 1M | $2.00 | $10.00 | 2026-09-28 | proprietary | |
| 3 | 0.660 | 0.703 | 1.0M | $10.00 | $50.00 | 2026-06-09 | proprietary | |
| 4 | 0.656 | 0.771 | 1M | $4.00 | $20.00 | 2026-09-22 | proprietary | |
| 5 | 0.655 | – | 1.05M | $0.10 | $0.50 | 2026-09-22 | proprietary | |
| 6 | 0.648 | 0.710 | 1M | $0.10 | $0.50 | 2026-10-07 | proprietary | |
| 7 | 0.647 | 0.682 | 1.05M | $10.00 | $50.00 | 2026-09-03 | proprietary | |
| 8 | 0.642 | 0.672 | 1.05M | $2.00 | $10.00 | 2026-09-29 | proprietary | |
| 9 | 0.633 | 0.689 | 1.0M | $10.00 | $50.00 | 2026-06-09 | proprietary | |
| 10 | 0.627 | 0.642 | 1.05M | $1.00 | $5.00 | 2026-10-26 | proprietary | |
| 11 | 0.623 | – | 1M | – | – | 2026-10-12 | proprietary | |
| 12 | 0.621 | 0.742 | 1.0M | $10.00 | $50.00 | 2026-09-01 | proprietary | |
| 13 | 0.608 | 0.612 | 1.05M | $0.10 | $0.50 | 2026-09-22 | proprietary | |
| 14 | 0.608 | 0.595 | 1.05M | $2.00 | $10.00 | 2026-09-22 | proprietary | |
| 15 | 0.605 | 0.641 | 1.05M | $4.00 | $20.00 | 2026-07-09 | proprietary | |
| 16 | 0.605 | – | 1M | – | – | 2026-10-12 | proprietary | |
| 17 | 0.598 | 0.734 | 1.0M | $5.00 | $25.00 | 2026-07-24 | proprietary | |
| 18 | 0.593 | – | 1.0M | $1.25 | $4.25 | 2026-07-09 | proprietary | |
| 19 | 0.578 | 0.624 | 1.0M | $2.00 | $10.00 | 2026-06-30 | proprietary | |
| 20 | 0.577 | 0.624 | 1.05M | $2.00 | $12.00 | 2026-07-09 | proprietary | |
| 21 | 0.574 | – | 1.0M | $5.00 | $25.00 | 2026-05-28 | proprietary | |
| 22 | 0.567 | – | 500K | $2.00 | $6.00 | 2026-09-21 | proprietary | |
| 23 | 0.558 | – | 1.0M | $5.00 | $25.00 | 2026-05-28 | proprietary | |
| 24 | 0.557 | 0.598 | 1.05M | $0.20 | $1.20 | 2026-07-09 | proprietary | |
| 25 | 0.551 | 0.642 | – | $2.00 | $10.00 | 2026-09-30 | proprietary | |
| 26 | 0.541 | – | 1.0M | – | – | 2026-04-08 | proprietary | |
| 27 | 0.540 | 0.566 | 1.05M | $4.00 | $20.00 | 2026-07-09 | proprietary | |
| 28 | 0.532 | 0.569 | 1M | $3.00 | $15.00 | 2026-07-16 | Kimi K3 License (open weights) | |
| 29 | 0.528 | 0.579 | 1M | $1.40 | $4.40 | 2026-08-18 | open | |
| 30 | 0.519 | – | 1M | $5.00 | $25.00 | 2026-04-16 | proprietary | |
| 31 | 0.518 | 0.572 | 1.05M | $5.00 | $30.00 | 2026-04-23 | proprietary | |
| 32 | 0.517 | 0.590 | 1M | $0.75 | $3.75 | 2026-09-02 | proprietary | |
| 33 | 0.507 | 0.529 | 1M | $0.30 | $1.20 | 2026-09-10 | open | |
| 34 | 0.485 | – | 500K | $2.00 | $6.00 | 2026-08-12 | proprietary | |
| 35 | 0.481 | 0.519 | 1.05M | $2.50 | $15.00 | 2026-03-05 | proprietary | |
| 36 | 0.462 | 0.510 | 400K | $1.25 | $10.00 | 2025-08-07 | proprietary | |
| 37 | 0.459 | 0.500 | 400K | $1.75 | $14.00 | 2025-12-11 | proprietary | |
| 38 | 0.442 | – | 1M | $3.00 | $15.00 | 2026-02-17 | proprietary | |
| 39 | 0.441 | 0.468 | 1.05M | $0.20 | $1.20 | 2026-07-09 | proprietary | |
| 40 | 0.396 | 0.480 | 400K | $1.25 | $10.00 | 2025-11-12 | proprietary | |
| 41 | 0.384 | 0.407 | 400K | $5.00 | $30.00 | 2026-05-05 | proprietary | |
| 42 | 0.350 | – | 256K | $2.00 | $8.00 | 2026-08-12 | proprietary | |
| 43 | 0.322 | – | 200K | $1.00 | $5.00 | 2025-10-15 | proprietary |
What the benchmark measures
HealthBench Professional evaluates language models on the work clinicians bring to a model: care consults such as differential diagnosis and management questions, clinical writing and documentation, and medical research. The 525 tasks were selected by physicians from 15,079 real clinician conversations, difficult cases were deliberately overweighted, and roughly a third of the set is adversarial.
190 physicians across 50 countries and 26 specialties wrote and adjudicated the rubrics. The full construction is on the benchmark page.
How to read the scores
A score is earned rubric points divided by possible points, clipped to 0 to 1. The length adjustment subtracts credit from answers longer than 2,000 characters and adds it to shorter ones, so verbosity cannot buy credit. Absolute numbers run low by design: the best result is 0.703, and physician-written responses score 0.437.
Scores depend on the grader version, the reasoning effort and the deployment setting a document reports, so 2 numbers from different documents can differ for reasons other than model quality. How each one is read is on the methodology page.
Data
The full results are published as JSON and CSV at stable URLs, licensed CC BY 4.0. Every score links to the document it was copied from on the sources page; citation format and version history are on the data page.