Inference
EN
Meta

Muse Spark 1.1

Muse Spark 1.1 is a reasoning model from Meta in the Muse Spark family. 38 benchmarks count toward its index, in 7 categories.

  • Closed weights
  • Reasoning
  • Input: text · image · video · file
  • Released 2026-07
RankedEnough results for an index.
IndexAll categories, out of 100.64.6±3.1Median 51.0
CoverageIndex weight that has results.95%
SpeedOutput tokens per second.152/sMedian 67/s
First tokenSeconds to the first output token.2.1 sMedian 1.3 s
Input / 1MUS dollars per 1M input tokens.$1.25Median $0.60
Output / 1MUS dollars per 1M output tokens.$4.25Median $2.55
ContextMaximum tokens in one request.1.05MMedian 262K
EloLMArena rating and rank.1479±4 #23Median 1422

Details

From OpenRouter
API IDThe model ID on OpenRouter.
meta/muse-spark-1.1
Max outputMaximum output tokens per request.
944K
Knowledge cutoffLast date of training data.
—
ToolsTool calls on OpenRouter.
Yes
JSON outputOutput follows a given JSON schema.
Yes
Cache read / 1MUS dollars per 1M cached tokens.
$0.15
WeightsLink to the published weights.
—
RetiresDate OpenRouter removes the model.
—

Where to run

APIs that serve this model.
2 providers
ProviderTypeMaker, cloud, host or router.InputOutputContext
MetaMaker$1.25$4.251.05M
OpenRouterRouter$1.25$4.251.05M

US dollars per 1M tokens, list price. From models.dev (opens in a new tab) (MIT).

Capabilities

Index per category, out of 100.
Out of 100
AgenticMulti-step tasks with tools.14 counted61.960 / 218
CodingCode writing and repair.8 counted64.843 / 209
ReasoningLogic problems and puzzles.4 counted63.054 / 227
MultimodalTasks with images and text.3 counted66.521 / 84
KnowledgeFacts and expert knowledge.7 counted67.433 / 268
MultilingualTasks in many languages.None counted——
InstructionTasks with strict prompt rules.1 counted67.5—
MathMath problems.1 counted61.2—

Results

38 counted
BenchmarkThe benchmark name.ResultThe score from the source.PlacePlace among results on this benchmark.
MCP AtlasAgentic88.1%2 / 52
ToolathlonAgenticAbove fitFar above this model's index.75.6%1 / 22
Humanity's Last ExamKnowledgeAbove fitFar above this model's index.62.1%6 / 77
Artificial Analysis Omniscience AccuracyKnowledge52.1%14 / 164
Artificial Analysis Humanity's Last ExamKnowledge46.2%24 / 222
Artificial Analysis SciCodeCoding58.8%14 / 114
MMLU ProKnowledge88.7% ±0.317 / 135
Humanity's Last Exam without toolsKnowledge52.2%7 / 53
GPQA Diamond · Vals AIKnowledge91.2% ±2.123 / 135
LiveCodeBenchCoding85.9% ±1.025 / 142
Artificial Analysis GPQA DiamondKnowledge89.8%38 / 199
MMMU ProMultimodal86.6% ±0.818 / 92
SimpleQA VerifiedKnowledge57.8% ±1.620 / 91
Critical Physics TasksReasoning15.1%50 / 214
SWE-benchCoding82.0% ±1.721 / 87
τ²-bench BankingAgentic40.5%7 / 28
SkillsBenchCoding59.2% ±4.49 / 35
CharXiv ReasoningMultimodal88.4%13 / 49
OSWorld-VerifiedAgentic80.8%10 / 37
BabyVisionMultimodal76.3%3 / 11
LiveBench Agentic CodingAgentic58.5%18 / 66
SWE-bench ProCoding61.5%28 / 83
LiveBench ReasoningReasoning87.7%27 / 80
Artificial Analysis Long Context ReasoningReasoning77.7%74 / 217
JobBenchAgentic54.7%13 / 36
Vibe Code Bench v1.1Coding72.2% ±3.346 / 123
GDPval-AA normalizedAgentic35.6%50 / 133
Terminal-Bench 2.1 · Vals AIAgentic69.3% ±1.030 / 75
Terminal-Bench 2.1 · BenchLMAgentic80.0%37 / 86
LiveBench Instruction FollowingInstruction69.6%30 / 66
DeepSearchQAAgentic84.9%12 / 24
Code MigrationCoding31.1% ±4.139 / 77
Chess PuzzlesReasoning34.0% ±4.847 / 90
LiveBench CodingCoding77.2%37 / 66
LiveBench Data AnalysisReasoning72.5%48 / 80
LiveBench MathematicsMath87.1%44 / 66
HealthBench ProfessionalKnowledge59.3%13 / 18
Agent Arena task outcomeAgentic−6.540 / 54
LiveBench LanguageKnowledge74.3%50 / 66
OSWorld 2.0Agentic14.2%19 / 25
Agent Arena command recoveryAgentic−2.338 / 50
DeepSWEAgentic53.3%49 / 59
CyberGymAgentic59.0%26 / 29
Agent Arena steerabilityAgentic−15.449 / 54
ExploitGymAgentic0.8%23 / 23
TaxEval v2Finance79.7% ±0.82 / 143
CorpFin v2Finance71.3% ±0.94 / 131
MedScribeHealthcare88.9% ±1.99 / 109
LegalBenchLegal85.0% ±0.427 / 149
Finance Agent (v2)Finance57.2% ±0.817 / 79
MortgageTaxFinance66.1% ±0.937 / 97
Legal Research BenchLegal38.0% ±3.443 / 90
ProgramBenchCoding0.0%34 / 65
Public Benefits Bench v1.1Public benefits61.3% ±1.330 / 50
EMBFinance56.4% ±3.245 / 75
Time Horizon Index: KSPOther1.7%13 / 13
CybenchAgentic92.9%—
WebArena-Verified Browser Agent BenchmarkAgentic69.0%—
MRCR 1MReasoning54.1%—

Sources

BenchLMCC BY-NC 4.0OpenRouterTerms restrict reuseVals AINo licence statedLiveBenchNo licence statedEpoch AICC BYSierra ResearchMITLMArenaCC BY 4.0models.devMIT

History

5 releases
Index65.9 → 64.6
23 Oct 202612 Nov 2026
Output / 1M$4.25 → $4.25
23 Oct 202612 Nov 2026

Questions

How much does Muse Spark 1.1 cost?
$1.25 per 1M input tokens, $4.25 per 1M output tokens.
How good is Muse Spark 1.1?
Its index is 64.6 out of 100. That is place 48 of 283 ranked models.
What is Muse Spark 1.1 best at?
Knowledge: 67.4 out of 100, place 33 of 268.
How fast is Muse Spark 1.1?
152 output tokens per second, 2.1 s to the first token.
What is the context window of Muse Spark 1.1?
1.05M tokens.
Is Muse Spark 1.1 open source?
No. Its weights are not published.
When was Muse Spark 1.1 released?
July 2026.

Badge

Muse Spark 1.1 on Inference360: place 48 of 283, index 64.6
HTML
<a href="https://inference360.ai/models/muse-spark-1-1"><img src="https://inference360.ai/badges/muse-spark-1-1.svg" alt="Muse Spark 1.1 on Inference360: place 48 of 283, index 64.6" height="20"></a>
Markdown
[![Muse Spark 1.1 on Inference360: place 48 of 283, index 64.6](https://inference360.ai/badges/muse-spark-1-1.svg)](https://inference360.ai/models/muse-spark-1-1)

Compare with

GPT-5.464.8Muse Ember64.8Muse Spark 1.264.9Qwen3.8-Flash-Next64.3MiMo-V2.6-Flash64.3

Lower price

dots3-note Preview63.4 · FreeGLM-5.3-Flash63.1 · $0.078/1MLing 3.1 Flash67.2 · $0.135/1M

More from Meta

All models
Muse Spark 1.264.9Muse Glimmer 30B53.0Muse Spark 1.2 ContributorUnrankedMuse Spark 1.370.1Muse Spark 1.3 ContributorUnranked