SShogo Bench

Muse Spark 1.1

Muse Spark is a natively multimodal reasoning model with support for tool-use, visual chain of thought, and multi-agent orchestration.

Shogo Agent Index

Estimated#40 of 92
48.3range 39.3–57.2 · 20% of weight measured
Agentic coding
53
17% measured
Long-horizon autonomy
64
30% measured
Knowledge-work agents
34
32% measured
Computer use
36
estimated
Security & context learning
48
estimated

Price

Input
$1.25/M
Output
$4.25/M
Cache read
$0.15/M
3:1 blend
$2.00/M
Agent-effective
$0.50/M

Cheapest host for agents: Meta at $0.50/M.

Agent-effective = 20 input tokens per output token, 85% cache hits. Why.

Agentic benchmarks

BenchmarkScoreNormalisedIndex weightHarness / source
SWE-bench Verifiedest.8%predicted 78
Terminal-Bench 2.0est.8%predicted 72
DeepSWE53.3%6%mini-swe-agent · $2.36/task
FrontierSWEest.5%predicted 16
CursorBenchest.4%predicted 38
GSO-Benchest.4%predicted 29
METR Time Horizonsest.10%predicted 75
Vending-Bench 2$6,5206%source
PostTrainBenchest.4%predicted 25
APEX-Agents31.8%8%source
GDPvalest.7%predicted 47
Remote Labor Indexest.5%predicted 4
DeepResearch Benchest.5%predicted 48
OSWorld 2.0est.6%predicted 10
OSWorldest.4%predicted 76
Cybenchest.5%predicted 75
CL-benchest.5%predicted 22

Other benchmarks

BenchmarkCategoryScoreSource
Chess PuzzlesReasoning34.0%source
CritPtReasoning15.1%source
DTBenchKnowledge94.4%source
LMCAKnowledge49.9%source
ProofBenchMath39.0%source
SciCodeCoding58.8%source
SimpleQA VerifiedKnowledge57.8%source
WebDev ArenaCoding1542source

API hosts

Live prices and uptime for every host serving Muse Spark 1.1 via OpenRouter.

HostInputOutputCache readAgent $/MUptime 24hQuant
Meta$1.25$4.25$0.15$0.50100.0%—