Inference
EN
Meta

Muse Spark 1.3

Muse Spark 1.3 is a reasoning model from Meta in the Muse Spark family. 41 benchmarks count toward its index, in 6 categories.

  • Closed weights
  • Reasoning
  • Input: text · image · video · file
  • Released 2026-09
RankedEnough results for an index.
IndexAll categories, out of 100.70.1±4.0Median 51.0
CoverageIndex weight that has results.85%
SpeedOutput tokens per second.76/sMedian 67/s
First tokenSeconds to the first output token.3.4 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.1491±5 #13Median 1422

Details

From OpenRouter
API IDThe model ID on OpenRouter.
meta/muse-spark-1.3
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.16 counted70.424 / 218
CodingCode writing and repair.10 counted68.823 / 209
ReasoningLogic problems and puzzles.5 counted68.130 / 227
MultimodalTasks with images and text.None counted——
KnowledgeFacts and expert knowledge.4 counted67.532 / 268
MultilingualTasks in many languages.None counted——
InstructionTasks with strict prompt rules.1 counted84.4—
MathMath problems.5 counted68.620 / 168

Results

41 counted
BenchmarkThe benchmark name.ResultThe score from the source.PlacePlace among results on this benchmark.
Artificial Analysis GPQA DiamondKnowledge93.5%8 / 199
LiveBench Instruction FollowingInstruction78.0%4 / 66
Artificial Analysis Humanity's Last ExamKnowledge48.7%15 / 222
GDPval-AA normalizedAgentic59.0%10 / 133
Terminal-Bench 2.1 · BenchLMAgentic88.8%7 / 86
Artificial Analysis Long Context ReasoningReasoning83.0%18 / 217
Artificial Analysis Tau3-BankingAgentic50.5%2 / 22
OTIS Mock AIME 2024-2025Math99.2% ±0.619 / 200
Critical Physics TasksReasoning24.9%21 / 214
DeepSWEAgentic75.4%6 / 59
LiveBench Agentic CodingAgentic64.1%8 / 66
Artificial Analysis SciCodeCoding58.8%14 / 114
JobBenchAgentic64.9%5 / 36
LiveBench MathematicsMath95.9%10 / 66
LiveBench Data AnalysisReasoning79.6%13 / 80
Vibe Code Bench v1.1Coding82.9% ±2.923 / 123
LiveBench ReasoningReasoning89.7%15 / 80
OSWorld 2.0Agentic66.9%5 / 25
LiveBench CodingCoding81.1%15 / 66
LiveBench Coding 2.0Coding90.2%13 / 52
Artificial Analysis Omniscience AccuracyKnowledge43.6%41 / 164
Agent Arena command recoveryAgentic6.113 / 50
FrontierMath-Tiers-1-3-v2-PrivateMath74.0% ±2.625 / 90
LiveBench LanguageKnowledge82.8%21 / 66
Terminal-Bench 2.1 · Vals AIAgentic72.3% ±0.424 / 75
FrontierMath-Tier-4-v2-PrivateMath46.3% ±7.924 / 72
LiveBench Browser UseAgentic80.0%17 / 49
MirrorCodeCoding54.4% ±10.814 / 39
Agent Arena task outcomeAgentic3.720 / 54
Agent Arena steerabilityAgentic2.622 / 54
LiveBench Long AgenticAgentic40.4%21 / 51
Chess PuzzlesReasoning37.0% ±4.939 / 90
AutomationBenchAgentic49.4%14 / 32
ApprenticeBench: end-to-end computer use, continual learning, and long-horizon agency on a real accounts-payable jobAgentic19.0%8 / 18
DeepSearchQAAgentic89.4%11 / 24
Mystery Game PuzzlesReasoning25.0% ±4.537 / 80
MazeBench: visual spatial reasoning in a 3D open worldAgentic0.0%23 / 48
Bug Hunt BenchCoding32.2%12 / 25
ProofBench v1.1Math55.0% ±5.030 / 54
Code MigrationCoding27.6% ±4.144 / 77
LiveBench DesignAgentic74.8%30 / 52
CWE-bench v1Agentic55.0%8 / 12
cursorBench40Coding41.6%14 / 21
FrontierSWE v2Coding25.8%15 / 22
LiveBench ResearchAgentic68.6%36 / 52
IOICoding43.9% ±1.433 / 45
Terminal-Bench 4.0Agentic10.6% ±1.845 / 61
Terminal-Bench 4.1.0Agentic14.8% ±1.520 / 23
Terminal-Bench 4.0.0Agentic14.6% ±1.520 / 23
Finance Agent (v2)Finance58.9% ±0.48 / 79
Tax Agent BenchFinance37.9% ±3.514 / 82
ProgramBenchCoding0.5% ±0.519 / 65
Legal Research BenchLegal40.9% ±3.433 / 90
EMBFinance62.7% ±3.031 / 75
CyberBench v1.1Other69.4% ±5.827 / 48
Terminal-Bench-ScienceScience4.3% ±2.426 / 44
OpenAI MRCR v2 8-needle 256K-512KReasoning98.5%—
OpenAI MRCR v2 8-needle 512K-1MReasoning98.1%—
SWE-Atlas Codebase QnACoding59.4%—

Sources

BenchLMCC BY-NC 4.0OpenRouterTerms restrict reuseEpoch AICC BYLiveBenchNo licence statedVals AINo licence statedTerminal-BenchNo licence statedLMArenaCC BY 4.0models.devMIT

History

5 releases
Index71.7 → 70.1
23 Oct 202612 Nov 2026
Output / 1M$4.25 → $4.25
23 Oct 202612 Nov 2026

Questions

How much does Muse Spark 1.3 cost?
$1.25 per 1M input tokens, $4.25 per 1M output tokens.
How good is Muse Spark 1.3?
Its index is 70.1 out of 100. That is place 22 of 283 ranked models.
What is Muse Spark 1.3 best at?
Coding: 68.8 out of 100, place 23 of 209.
How fast is Muse Spark 1.3?
76 output tokens per second, 3.4 s to the first token.
What is the context window of Muse Spark 1.3?
1.05M tokens.
Is Muse Spark 1.3 open source?
No. Its weights are not published.
When was Muse Spark 1.3 released?
September 2026.

Badge

Muse Spark 1.3 on Inference360: place 22 of 283, index 70.1
HTML
<a href="https://inference360.ai/models/muse-spark-1-3"><img src="https://inference360.ai/badges/muse-spark-1-3.svg" alt="Muse Spark 1.3 on Inference360: place 22 of 283, index 70.1" height="20"></a>
Markdown
[![Muse Spark 1.3 on Inference360: place 22 of 283, index 70.1](https://inference360.ai/badges/muse-spark-1-3.svg)](https://inference360.ai/models/muse-spark-1-3)

Compare with

MiMo-V2.6-Pro70.2MiniMax M3.1 Flash69.3Grok 4.769.1GPT-5.569.1GPT-5.6 Terra68.9

Lower price

Ling 3.1 Flash67.2 · $0.135/1MDeepSeek V4.1 Flash 101768.5 · $0.40/1MMiniMax M3.1 Flash69.3 · $0.60/1M

More from Meta

All models
Muse Spark 1.3 ContributorUnrankedMuse Spark 1.2 ContributorUnrankedMuse Glimmer 30B53.0Muse Spark 1.264.9Muse Wonder Full77.3