Muse Spark

Meta · meta/muse-spark

ProviderMeta
Familymuse
Typellm-reasoning
Statusactive
Released2026-04-08
Updated2026-04-08

Reported benchmark scores

These scores carry one collection date for the whole card (2026-04) and a source list rather than a source per score, so they cannot be attributed individually. They are shown as reported and are not verified evidence. Benchmark pages carry the reviewed, dated evidence where it exists.
BenchmarkCatalogue standingScore
ARC-AGI-2unassessed42.5
Artificial Analysis Intelligence Indexunassessed52.0
CharXiv Reasoningunassessed86.4
DeepSearchQAunassessed74.8
FrontierScience — Research trackunassessed38.3
HealthBench Hardunassessed42.8
Humanity's Last Examunassessed50.2
IPhO 2025 Theoryunassessed82.6
MedXpertQA MMunassessed78.4
Terminal-Bench 2.0unassessed59.0
ZeroBenchunassessed33.0

Data

This card as JSON · Edit on GitHub