Models

31 models ranked · updated September 30, 2026

Every model on the HealthBench Professional leaderboard, with the API facts that decide whether a score is usable in practice: context window, per-token price, and license. Each model has its own page with the full result, the document its score was read from, and head-to-head comparisons.

#	model		score	size	context	cost in / out per 1M	license
1		GPT-6 Astra (Anthropic run) OpenAI via Claude Sonnet 5.5 System Card	0.703	—	1.05M	$10.00 / $50.00	proprietary
2		Claude Sonnet 5.5 Anthropic via Claude Sonnet 5.5 System Card	0.692	—	1M	$2.00 / $10.00	proprietary
3		Claude Fable 5 Anthropic via Claude Fable 5 and Claude Mythos 5 System Card	0.660	—	1.0M	$10.00 / $50.00	proprietary
4		Claude Opus 5.5 Anthropic via Claude Opus 5.5 System Card	0.656	—	1M	$4.00 / $20.00	proprietary
5		GPT-6 Astra OpenAI via GPT-6 Astra System Card - HealthBench (Deployment Safety Hub)	0.647	—	1.05M	$10.00 / $50.00	proprietary
6		Claude Fable 5 (September card) Anthropic via Claude Fable 5.1 and Claude Mythos 5.1 System Card	0.633	—	1.0M	$10.00 / $50.00	proprietary
7		Claude Fable 5.1 Anthropic via Claude Fable 5.1 and Claude Mythos 5.1 System Card	0.621	—	1.0M	$10.00 / $50.00	proprietary
8		GPT-6 Luna OpenAI via GPT-6 Astra System Card - HealthBench (Deployment Safety Hub)	0.608	—	1.05M	$0.10 / $0.50	proprietary
9		GPT-6 Sol OpenAI via GPT-6 Astra System Card - HealthBench (Deployment Safety Hub)	0.608	—	1.05M	$2.00 / $10.00	proprietary
10		GPT-5.6 Sol OpenAI via GPT-5.6 System Card	0.605	—	1.05M	$4.00 / $20.00	proprietary
11		Claude Opus 5 Anthropic via System Card: Claude Opus 5	0.598	—	1.0M	$5.00 / $25.00	proprietary
12		Muse Spark 1.1 Meta via Muse Spark 1.1 Evaluation Report	0.593	—	1.0M	$1.25 / $4.25	proprietary
13		Claude Sonnet 5 Anthropic via System Card: Claude Sonnet 5	0.578	—	1.0M	$2.00 / $10.00	proprietary
14		GPT-5.6 Terra OpenAI via GPT-5.6 System Card	0.577	—	1.05M	$2.00 / $12.00	proprietary
15		Claude Opus 4.8 (Opus 4.8 grader) Anthropic via System Card: Claude Sonnet 5	0.574	—	1.0M	$5.00 / $25.00	proprietary
16		Grok 4.7 SpaceX AI via Introducing Grok 4.7	0.567	—	500K	$2.00 / $6.00	proprietary
17		Claude Opus 4.8 Anthropic via System Card: Claude Opus 4.8	0.558	—	1.0M	$5.00 / $25.00	proprietary
18		GPT-5.6 Luna OpenAI via GPT-5.6 System Card	0.557	—	1.05M	$0.20 / $1.20	proprietary
19		Muse Spark Meta via Muse Spark 1.1 Evaluation Report	0.541	—	1.0M	—	proprietary
20		GPT-5.6 Sol (August) OpenAI via GPT-5.6 - August Updates (system card addendum)	0.540	—	1.05M	$4.00 / $20.00	proprietary
21		Claude Opus 4.7 Anthropic via System Card: Claude Opus 4.8	0.519	—	1M	$5.00 / $25.00	proprietary
22		GPT-5.5 OpenAI via GPT-5.6 System Card	0.518	—	1.05M	$5.00 / $30.00	proprietary
23		Grok 4.6 SpaceX AI via Introducing Grok 4.7	0.485	—	500K	$2.00 / $6.00	proprietary
24		GPT-5.4 OpenAI via GPT-5.6 System Card	0.481	—	1.05M	$2.50 / $15.00	proprietary
25		GPT-5 OpenAI via GPT-5.6 System Card	0.462	—	400K	$1.25 / $10.00	proprietary
26		GPT-5.2 OpenAI via GPT-5.6 System Card	0.459	—	400K	$1.75 / $14.00	proprietary
27		Claude Sonnet 4.6 Anthropic via System Card: Claude Sonnet 5	0.442	—	1M	$3.00 / $15.00	proprietary
28		GPT-5.6 Luna (August) OpenAI via GPT-5.6 - August Updates (system card addendum)	0.441	—	1.05M	$0.20 / $1.20	proprietary
29		GPT-5.1 OpenAI via GPT-5.6 System Card	0.396	—	400K	$1.25 / $10.00	proprietary
30		GPT-5.5 Instant OpenAI via GPT-5.5 Instant System Card	0.384	—	400K	$5.00 / $30.00	proprietary
31		MAI-Thinking-1 Microsoft via MAI-Thinking-1: Building a Hill-Climbing Machine	0.350	1.0T	256K	$2.00 / $8.00	proprietary

Scores are compiled from published documents; the line under each model name names the document and links to the sources page. Rows marked source pending have no located document yet.

Model pages

• GPT-6 Astra (Anthropic run) on HealthBench Professional
  rank 1 of 31 · score 0.703 · OpenAI
• Claude Sonnet 5.5 on HealthBench Professional
  rank 2 of 31 · score 0.692 · Anthropic
• Claude Fable 5 on HealthBench Professional
  rank 3 of 31 · score 0.660 · Anthropic
• Claude Opus 5.5 on HealthBench Professional
  rank 4 of 31 · score 0.656 · Anthropic
• GPT-6 Astra on HealthBench Professional
  rank 5 of 31 · score 0.647 · OpenAI
• Claude Fable 5 (September card) on HealthBench Professional
  rank 6 of 31 · score 0.633 · Anthropic
• Claude Fable 5.1 on HealthBench Professional
  rank 7 of 31 · score 0.621 · Anthropic
• GPT-6 Luna on HealthBench Professional
  rank 8 of 31 · score 0.608 · OpenAI
• GPT-6 Sol on HealthBench Professional
  rank 9 of 31 · score 0.608 · OpenAI
• GPT-5.6 Sol on HealthBench Professional
  rank 10 of 31 · score 0.605 · OpenAI
• Claude Opus 5 on HealthBench Professional
  rank 11 of 31 · score 0.598 · Anthropic
• Muse Spark 1.1 on HealthBench Professional
  rank 12 of 31 · score 0.593 · Meta
• Claude Sonnet 5 on HealthBench Professional
  rank 13 of 31 · score 0.578 · Anthropic
• GPT-5.6 Terra on HealthBench Professional
  rank 14 of 31 · score 0.577 · OpenAI
• Claude Opus 4.8 (Opus 4.8 grader) on HealthBench Professional
  rank 15 of 31 · score 0.574 · Anthropic
• Grok 4.7 on HealthBench Professional
  rank 16 of 31 · score 0.567 · SpaceX AI
• Claude Opus 4.8 on HealthBench Professional
  rank 17 of 31 · score 0.558 · Anthropic
• GPT-5.6 Luna on HealthBench Professional
  rank 18 of 31 · score 0.557 · OpenAI
• Muse Spark on HealthBench Professional
  rank 19 of 31 · score 0.541 · Meta
• GPT-5.6 Sol (August) on HealthBench Professional
  rank 20 of 31 · score 0.540 · OpenAI
• Claude Opus 4.7 on HealthBench Professional
  rank 21 of 31 · score 0.519 · Anthropic
• GPT-5.5 on HealthBench Professional
  rank 22 of 31 · score 0.518 · OpenAI
• Grok 4.6 on HealthBench Professional
  rank 23 of 31 · score 0.485 · SpaceX AI
• GPT-5.4 on HealthBench Professional
  rank 24 of 31 · score 0.481 · OpenAI
• GPT-5 on HealthBench Professional
  rank 25 of 31 · score 0.462 · OpenAI
• GPT-5.2 on HealthBench Professional
  rank 26 of 31 · score 0.459 · OpenAI
• Claude Sonnet 4.6 on HealthBench Professional
  rank 27 of 31 · score 0.442 · Anthropic
• GPT-5.6 Luna (August) on HealthBench Professional
  rank 28 of 31 · score 0.441 · OpenAI
• GPT-5.1 on HealthBench Professional
  rank 29 of 31 · score 0.396 · OpenAI
• GPT-5.5 Instant on HealthBench Professional
  rank 30 of 31 · score 0.384 · OpenAI
• MAI-Thinking-1 on HealthBench Professional
  rank 31 of 31 · score 0.350 · Microsoft