Muse Spark on HealthBench Professional

rank 11 of 22 · updated September 8, 2026

Muse Spark scores 0.541 on HealthBench Professional, rank 11 of 22 models on the board. Meta's April 2026 frontier model, the first of the Muse Spark line, with the lowest list price on this board. 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	11 of 22
score	0.541
lab	Meta
context window	1.0M tokens
API price per 1M tokens	$1.25 in / $4.25 out
license	proprietary
source	Muse Spark 1.1 Evaluation Report (model card)
released	2026-04-08

Position in the field

The gap to the leader, Claude Fable 5 at 0.660, is 0.119. Directly above sits GPT-5.6 Luna at 0.557. Directly below sits GPT-5.6 Sol (August) at 0.540. 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 Muse Spark 1.1 Evaluation Report (model card, Meta, 2026-07-09). Vendor-reported. Confidence: verified. Configuration: length-normalized, GPT-5.4 low-reasoning grader; Muse Spark (1.0) column in the Muse Spark 1.1 Evaluation Report Figure 44.

Health | HealthBench Professional | 59.3 | 54.1 | 41.6 | 55.8 | 51.8 (Figure 44 image table row; columns Muse Spark 1.1, Muse Spark, Gemini 3.1 Pro (high), Opus 4.8 (max), GPT 5.5 (xhigh))

p. 101, Figure 44 (image), row HealthBench Professional, column Muse Spark; protocol p. 104 (printed 103) · full entry on the sources page

What does Muse Spark score on HealthBench Professional?

Muse Spark scores 0.541 on HealthBench Professional, which places it at rank 11 of 22 models on the board as of September 8, 2026. The number was read from Muse Spark 1.1 Evaluation Report, listed on the sources page.

How much does Muse Spark cost to run?

Muse Spark is priced at $1.25 per million input tokens and $4.25 per million output tokens through Meta's API.

Head to head

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

Muse Spark vs Claude Fable 5
0.541 vs 0.660 · Claude Fable 5 by 0.119
Muse Spark vs GPT-6 Astra
0.541 vs 0.634 · GPT-6 Astra by 0.093
Muse Spark vs Claude Fable 5.1
0.541 vs 0.621 · Claude Fable 5.1 by 0.080
Muse Spark vs GPT-5.6 Sol
0.541 vs 0.605 · GPT-5.6 Sol by 0.064
Muse Spark vs Claude Opus 5
0.541 vs 0.598 · Claude Opus 5 by 0.057
Muse Spark vs Muse Spark 1.1
0.541 vs 0.593 · Muse Spark 1.1 by 0.052
Muse Spark vs Claude Sonnet 5
0.541 vs 0.578 · Claude Sonnet 5 by 0.037
Muse Spark vs GPT-5.6 Terra
0.541 vs 0.577 · GPT-5.6 Terra by 0.036
Muse Spark vs Claude Opus 4.8
0.541 vs 0.558 · Claude Opus 4.8 by 0.017
Muse Spark vs GPT-5.6 Luna
0.541 vs 0.557 · GPT-5.6 Luna by 0.016
Muse Spark vs GPT-5.6 Sol (August)
0.541 vs 0.540 · Muse Spark by 0.001
Muse Spark vs Claude Opus 4.7
0.541 vs 0.519 · Muse Spark by 0.022
Muse Spark vs GPT-5.5
0.541 vs 0.518 · Muse Spark by 0.023
Muse Spark vs GPT-5.4
0.541 vs 0.481 · Muse Spark by 0.060
Muse Spark vs GPT-5
0.541 vs 0.462 · Muse Spark by 0.079
Muse Spark vs GPT-5.2
0.541 vs 0.459 · Muse Spark by 0.082
Muse Spark vs Claude Sonnet 4.6
0.541 vs 0.442 · Muse Spark by 0.099
Muse Spark vs GPT-5.6 Luna (August)
0.541 vs 0.441 · Muse Spark by 0.100
Muse Spark vs GPT-5.1
0.541 vs 0.396 · Muse Spark by 0.145
Muse Spark vs GPT-5.5 Instant
0.541 vs 0.384 · Muse Spark by 0.157
Muse Spark vs MAI-Thinking-1
0.541 vs 0.350 · Muse Spark by 0.191

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