benchgap
Meta · model

Muse Spark benchmark scores

As of 2026-11-21, Muse Spark (Meta) has measured scores on 36 benchmarks and estimated scores on 112 more.

BenchmarkScoreSource
AA-GPQA Diamond88.4%measured
AA-HLE40.7%measured
AA-SciCode48.7%estimated ± 3.4 pp, high confidence
Artificial Analysis Intelligence Index31.3%measured
CritPt11.3%measured
GDPval-AA24.9%measured
GPQA Diamond89.5%measured
HLE50.4%measured
IFBench74.2%estimated ± 6.6 pp, medium confidence
AA-LCR78.0%measured
LiveCodeBench v686.6%estimated ± 2.7 pp, high confidence
AA-Omniscience Accuracy49.6%measured
SciCode46.4%estimated ± 3.6 pp, high confidence
BrowseComp60.1%estimated ± 5.2 pp, medium confidence
HLE w/ tools50.0%estimated ± 4.9 pp, medium confidence
IFEval93.0%estimated ± 2.8 pp, high confidence
LongBench v264.0%estimated ± 5.5 pp, medium confidence
VITA-Bench41.0%estimated ± 10.7 pp, low confidence
τ²-bench results91.5%measured
AA Coding Index58.6%measured
AA-MMMU-Pro80.5%measured
APEX-Agents18.0%estimated ± 9.4 pp, low confidence
APEX-Agents-AA14.0%estimated ± 6.0 pp, low confidence
BioMysteryBench (human-difficult)44.7%estimated ± 6.8 pp, low confidence
DeepSearchQA74.8%measured
SWE-bench Verified77.4%measured
BFCL v473.4%estimated ± 2.7 pp, high confidence
CyberGym43.5%measured
JobBench31.2%estimated ± 8.6 pp, low confidence
SWE-bench Pro52.4%measured
WideResearch70.1%estimated ± 1.6 pp, medium confidence
AA Agentic Index19.1%estimated ± 3.1 pp, high confidence
MMLU-Pro85.7%estimated ± 2.5 pp, low confidence
AA-IFBench75.9%measured
GPQA89.5%estimated ± 0.0 pp, medium confidence
MMLU90.5%estimated ± 0.9 pp, medium confidence
Vals GPQA Diamond89.6%measured
Vals LiveCodeBench79.7%estimated ± 4.7 pp, medium confidence
Vals MMLU-Pro87.3%measured
MMMLU89.7%estimated ± 0.7 pp, medium confidence
MMMU86.4%estimated ± 0.8 pp, medium confidence
AA MMLU-Pro88.8%estimated ± 0.7 pp, high confidence
Gert Labs54.8%estimated ± 5.5 pp, medium confidence
AA EnterpriseOps-Gym43.2%estimated ± 2.3 pp, medium confidence
AA Tau3 Banking29.9%estimated ± 4.1 pp, high confidence
ApprenticeBench8.4%estimated ± 10.1 pp, low confidence
ARC-AGI-182.0%estimated ± 1.7 pp, high confidence
ARC-AGI-242.5%measured
CursorBench 3.147.9%estimated ± 4.8 pp, low confidence
CursorBench 3.245.4%estimated ± 3.6 pp, medium confidence
FrontierCode 1.1 Main22.9%estimated ± 4.2 pp, low confidence
FrontierSWE v22.5%estimated ± 8.3 pp, low confidence
MazeBench0.4%estimated ± 5.5 pp, low confidence
OfficeQA Pro60.4%estimated ± 6.8 pp, medium confidence
OSWorld-Verified71.0%estimated ± 3.8 pp, medium confidence
ParseBench-161.9%estimated ± 5.2 pp, low confidence
Terminal-Bench 4.12.2%estimated ± 9.2 pp, low confidence
Terminal-Bench Hard45.5%measured
Vals SWE-bench74.4%measured
Terminal-Bench 2.1 (Vals AI)57.4%estimated ± 5.6 pp, medium confidence
CWE-Bench-AA32.4%estimated ± 10.1 pp, low confidence
DeepsecBench-AA1.2%estimated ± 8.9 pp, low confidence
HealthBench Professional (length-adjusted, Artificial Analysis)55.4%estimated ± 4.3 pp, low confidence
HealthBench Professional (raw, Artificial Analysis)57.2%estimated ± 0.8 pp, low confidence
AA ITBench37.3%estimated ± 5.1 pp, low confidence
Bug Hunt Bench7.4%estimated ± 9.0 pp, low confidence
Chartography (tools)83.7%estimated ± 4.1 pp, low confidence
CursorBench 4.029.8%estimated ± 3.3 pp, medium confidence
CWE-bench v134.3%estimated ± 6.4 pp, low confidence
DeepSWE30.0%estimated ± 6.4 pp, low confidence
DRACO72.4%estimated ± 2.0 pp, medium confidence
FrontierCode 2.019.3%estimated ± 6.7 pp, low confidence
HLE w/o tools42.8%measured
OSWorld 2.1 Offline2.4%estimated ± 1.9 pp, low confidence
PostTrainBench v1.119.3%estimated ± 3.7 pp, medium confidence
Toolathlon-Verified51.0%estimated ± 3.1 pp, medium confidence
VulcanBench Frontier v475.1%estimated ± 8.3 pp, low confidence
GDP.pdf17.0%estimated ± 4.6 pp, medium confidence
ARC-AGI-34.8%estimated ± 7.2 pp, medium confidence
ARC-AGI-3 Provider Adapter2.7%estimated ± 7.3 pp, medium confidence
MKA21.1%estimated ± 6.4 pp, low confidence
Vibe Code Bench19.7%measured
Hermes Bench55.7%estimated ± 2.9 pp, high confidence
Hermes Index22.8%estimated ± 8.7 pp, medium confidence
ParseBench-272.7%estimated ± 5.0 pp, medium confidence
CharXiv86.4%measured
CharXiv w/o tools84.6%estimated ± 1.2 pp, high confidence
ExploitGym1.5%estimated ± 9.3 pp, low confidence
AI2D_TEST93.5%estimated ± 0.9 pp, low confidence
C-Eval94.9%estimated ± 1.0 pp, low confidence
Claw-Eval63.8%measured
CountBench96.1%estimated ± 2.2 pp, medium confidence
DynaMath87.4%estimated ± 0.8 pp, low confidence
ERQA64.7%measured
MathVision100.0%estimated ± 2.2 pp, low confidence
MCP Atlas62.8%estimated ± 10.7 pp, low confidence
MCP-Tasks80.4%estimated ± 6.4 pp, low confidence
MLVU (M-Avg)87.8%estimated ± 1.5 pp, high confidence
MMLU-Redux97.2%estimated ± 0.8 pp, low confidence
MMMU-Pro80.4%measured
NL2Repo42.1%estimated ± 4.7 pp, high confidence
OmniDocBench 1.589.8%estimated ± 2.0 pp, medium confidence
PostTrainBench v119.0%estimated ± 3.1 pp, medium confidence
QwenClawBench55.9%estimated ± 6.0 pp, low confidence
RealWorldQA87.6%estimated ± 1.4 pp, medium confidence
ScreenSpot Pro84.1%measured
SimpleVQA71.3%measured
SuperGPQA89.6%estimated ± 7.2 pp, medium confidence
Terminal-Bench 2.059.0%measured
Toolathlon43.1%estimated ± 6.3 pp, medium confidence
V*96.1%estimated ± 6.5 pp, low confidence
Video-MME (w/o subtitle)85.7%estimated ± 1.3 pp, high confidence
Video-MME (with subtitle)89.3%estimated ± 2.9 pp, high confidence
VideoMMMU86.4%estimated ± 1.5 pp, medium confidence
We-Math83.8%estimated ± 2.1 pp, medium confidence
AA LiveCodeBench88.6%estimated ± 2.6 pp, high confidence
HealthBench Hard42.8%measured
LiveCodeBench Pro80.0%measured
MedXpertQA (MM)78.4%measured
MedXpertQA (Text)52.6%measured
MMLU-Pro (Arcee)86.7%estimated ± 3.4 pp, low confidence
React Native Evals78.5%estimated ± 2.7 pp, medium confidence
ResearchClawBench17.7%estimated ± 1.3 pp, high confidence
SWE-Rebench60.6%estimated ± 3.6 pp, medium confidence
PostTrainBench16.7%estimated ± 3.7 pp, medium confidence
BioMysteryBench (human-solvable)84.4%estimated ± 0.6 pp, low confidence
FrontierCode 1.1 Extended53.3%estimated ± 3.0 pp, low confidence
HealthBench (raw)60.3%estimated ± 0.7 pp, low confidence
HealthBench (length-adjusted)60.1%estimated ± 0.4 pp, low confidence
HealthBench Professional52.1%estimated ± 4.2 pp, medium confidence
HealthBench Professional (raw)74.3%estimated ± 1.7 pp, low confidence
HLE-Verified15.4%estimated ± 2.0 pp, low confidence
LABBench281.6%estimated ± 1.8 pp, low confidence
ScarfBench25.9%estimated ± 9.1 pp, low confidence
STEMBench2.6%estimated ± 0.2 pp, medium confidence
SimpleQA32.8%estimated ± 7.8 pp, low confidence
LiveCodeBench77.7%estimated ± 14.6 pp, low confidence
BabyVision w/ Python88.9%estimated ± 3.8 pp, low confidence
OpenHarmony Bench v1.051.4%estimated ± 2.8 pp, high confidence
BabyVision66.2%estimated ± 9.1 pp, low confidence
MMVU78.6%estimated ± 2.7 pp, medium confidence
PerceptionBench68.3%estimated ± 2.7 pp, medium confidence
ZeroBench33.0%measured
MRCRv285.5%estimated ± 11.3 pp, low confidence
MMLongBench-Doc61.3%estimated ± 2.6 pp, low confidence
OCRBench V271.7%estimated ± 3.3 pp, high confidence
BioBasicsBench34.0%estimated ± 2.8 pp, medium confidence
MMMU-Pro w/ Python83.2%estimated ± 2.0 pp, high confidence