Claude Sonnet 4.6 on HealthBench Professional

rank 18 of 22 · updated September 8, 2026

Claude Sonnet 4.6 scores 0.442 on HealthBench Professional, rank 18 of 22 models on the board. The February 2026 Sonnet, Anthropic's mid-tier model before the Claude 5 line. HealthBench Professional scores models on 525 tasks drawn from real clinician conversations, graded criterion by criterion against physician-written rubrics on a 0 to 1 scale.

Result and API facts

rank	18 of 22
score	0.442
lab	Anthropic
context window	1M tokens
API price per 1M tokens	$3.00 in / $15.00 out
license	proprietary
source	System Card: Claude Sonnet 5 (system card)
released	2026-02-17

Position in the field

The gap to the leader, Claude Fable 5 at 0.660, is 0.218. Directly above sits GPT-5.2 at 0.459. Directly below sits GPT-5.6 Luna (August) at 0.441. The scores on this page are compiled from published documents rather than from one controlled run, so a small gap between 2 models can reflect a difference in grader version, reasoning effort, or deployment setting as well as a difference in capability.

Source of this score

Read from System Card: Claude Sonnet 5 (system card, Anthropic, 2026-06-30). Vendor-reported. Confidence: verified. Configuration: length-adjusted; Anthropic protocol: adaptive thinking at max effort, Claude Opus 4.8 grader, averaged over 5 trials, no tools or custom system prompt; comparison column in the Sonnet 5 card. Other Anthropic prints: 44.4% (Fable card Figure 8.18.2.A), 41.7% (Opus 4.8 card, Sonnet 4.6 grader).

HealthBench Professional 57.8 44.2 51.8 -

p. 115, Table 8.1.A, row HealthBench Professional, column Claude Sonnet 4.6 · 1 corroborating document · full entry on the sources page

What does Claude Sonnet 4.6 score on HealthBench Professional?

Claude Sonnet 4.6 scores 0.442 on HealthBench Professional, which places it at rank 18 of 22 models on the board as of September 8, 2026. The number was read from System Card: Claude Sonnet 5, listed on the sources page.

How much does Claude Sonnet 4.6 cost to run?

Claude Sonnet 4.6 is priced at $3.00 per million input tokens and $15.00 per million output tokens through Anthropic's API.

Head to head

Pairings with a dedicated comparison page are linked; every other difference is in the score-difference matrix.

Claude Sonnet 4.6 vs Claude Fable 5
0.442 vs 0.660 · Claude Fable 5 by 0.218
Claude Sonnet 4.6 vs GPT-6 Astra
0.442 vs 0.634 · GPT-6 Astra by 0.192
Claude Sonnet 4.6 vs Claude Fable 5.1
0.442 vs 0.621 · Claude Fable 5.1 by 0.179
Claude Sonnet 4.6 vs GPT-5.6 Sol
0.442 vs 0.605 · GPT-5.6 Sol by 0.163
Claude Sonnet 4.6 vs Claude Opus 5
0.442 vs 0.598 · Claude Opus 5 by 0.156
Claude Sonnet 4.6 vs Muse Spark 1.1
0.442 vs 0.593 · Muse Spark 1.1 by 0.151
Claude Sonnet 4.6 vs Claude Sonnet 5
0.442 vs 0.578 · Claude Sonnet 5 by 0.136
Claude Sonnet 4.6 vs GPT-5.6 Terra
0.442 vs 0.577 · GPT-5.6 Terra by 0.135
Claude Sonnet 4.6 vs Claude Opus 4.8
0.442 vs 0.558 · Claude Opus 4.8 by 0.116
Claude Sonnet 4.6 vs GPT-5.6 Luna
0.442 vs 0.557 · GPT-5.6 Luna by 0.115
Claude Sonnet 4.6 vs Muse Spark
0.442 vs 0.541 · Muse Spark by 0.099
Claude Sonnet 4.6 vs GPT-5.6 Sol (August)
0.442 vs 0.540 · GPT-5.6 Sol (August) by 0.098
Claude Sonnet 4.6 vs Claude Opus 4.7
0.442 vs 0.519 · Claude Opus 4.7 by 0.077
Claude Sonnet 4.6 vs GPT-5.5
0.442 vs 0.518 · GPT-5.5 by 0.076
Claude Sonnet 4.6 vs GPT-5.4
0.442 vs 0.481 · GPT-5.4 by 0.039
Claude Sonnet 4.6 vs GPT-5
0.442 vs 0.462 · GPT-5 by 0.020
Claude Sonnet 4.6 vs GPT-5.2
0.442 vs 0.459 · GPT-5.2 by 0.017
Claude Sonnet 4.6 vs GPT-5.6 Luna (August)
0.442 vs 0.441 · Claude Sonnet 4.6 by 0.001
Claude Sonnet 4.6 vs GPT-5.1
0.442 vs 0.396 · Claude Sonnet 4.6 by 0.046
Claude Sonnet 4.6 vs GPT-5.5 Instant
0.442 vs 0.384 · Claude Sonnet 4.6 by 0.058
Claude Sonnet 4.6 vs MAI-Thinking-1
0.442 vs 0.350 · Claude Sonnet 4.6 by 0.092

Where the scores come from and how they are read is on the methodology page. The full ranking is on the leaderboard.