BenchLeader
Menu
Best LLMsChoose a modelLeaderboardsModelsBenchmarksProvidersReleases9

Muse Spark 1.1 vs Muse Spark 1.2

Full profiles: Muse Spark 1.1 (Meta) and Muse Spark 1.2 (Meta): every benchmark, price and rank.

Updated · Data from 9 sources

As of 25 Oct 2026, Muse Spark 1.1 and Muse Spark 1.2 (xhigh) are neck and neck on quality (62.5 vs 63.2, inside the ±7.2 margin of error) and cost about the same, so speed decides it: Muse Spark 1.2 (xhigh) streams 1.5× faster. Pick Muse Spark 1.1 for coding (66 vs 55); pick Muse Spark 1.2 for maths (67 vs 61).

Muse Spark 1.1 benchmarks Meta Muse Spark 1.2 benchmarks xhigh MetaOpen in the comparison tool

Which to pick for…

The pick's numbers come first. Category scores on the BenchLeader scale (50 is the average model, 15 points one standard deviation). High confidence means a lead of half a standard deviation or more; under one point is called either.

Use casePickWhyConfidence
CodingMuse Spark 1.166 vs 55 category score; Arena WebDev 1543 vs 1533High
Agents & tool useMuse Spark 1.255 vs 52 category score; Arena Agent -7 vs -6.9Medium
MathsMuse Spark 1.267 vs 61 category score; Arena Maths 1491 vs 1493Medium
ReasoningMuse Spark 1.269 vs 68 category score; Arena Hard Prompts 1512 vs 1510Medium
KnowledgeMuse Spark 1.169 vs 67 category score; SimpleQA Verified 57.8% vs 60.3%Medium
Images and other mediaMuse Spark 1.265 vs 64 category score; Arena Vision 1304 vs 1293Medium
Chat quality (human preference)Muse Spark 1.269 vs 65 on Arena votesMedium
Lowest costEither$2.00 vs $2.00 per 1M blended; agentic coding session $0.092 vs $0.092–
Fastest first answerMuse Spark 1.11.5 s (Muse Spark 1.1) vs 17.6 s (Muse Spark 1.2 (xhigh)) to the first answer token–
High volume on a budgetNeither flagshipA smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs–

Key differences

  • Muse Spark 1.1 and Muse Spark 1.2 (xhigh) are neck and neck on quality (62.5 vs 63.2, within the ±7.2 margin).
  • Muse Spark 1.1 is stronger in coding, instruction following, knowledge.
  • Muse Spark 1.2 (xhigh) is stronger in agents & tools, human preference, maths, multimodal, reasoning, composite, long context.
  • They cost about the same ($2.00 per 1M blended).
  • Muse Spark 1.2 (xhigh) streams 1.5× faster (307 vs 200 tokens per second).

At a glance

SpecMuse Spark 1.1Muse Spark 1.2
DeveloperMetaMeta
Released9 Jul 20265 Aug 2026
Context window (tokens)1,048,5761,048,576
Compared atMuse Spark 1.1Muse Spark 1.2 (xhigh)
Open weightsNoNo
Price in / out / cached input per 1M$1.25 / $4.25 / $0.150$1.25 / $4.25 / $0.150

Category by category

Category scores on the BenchLeader scale: 50 is the average evaluated model, 15 points is one standard deviation. A gap under one point is called level.

CategoryMuse Spark 1.1Muse Spark 1.2 (xhigh)Leader
Reasoning67.969.3Muse Spark 1.2
Coding66.355.1Muse Spark 1.1
Agents & tools51.855.4Muse Spark 1.2
Maths61.367.3Muse Spark 1.2
Knowledge69.266.7Muse Spark 1.1
Instruction following71.5–not comparable
Human preference65.068.6Muse Spark 1.2
Multimodal63.665.2Muse Spark 1.2
Long context–64.5not comparable
Composite–73.8not comparable

Where sources disagree

Across the 14 benchmarks both have, Muse Spark 1.1 leads 5, Muse Spark 1.2 leads 9. Every overall index that rates both puts Muse Spark 1.2 ahead.

Overall indexMuse Spark 1.1Muse Spark 1.2 (xhigh)Leader
BenchLeader Index62.563.2neck and neck
AA Intelligence Index v4.3.233.7†39.6Muse Spark 1.2
Arena Text14901491Muse Spark 1.2

Each publisher runs its own tasks, harnesses and effort settings, so their overall verdicts can differ; BenchLeader reconciles them on one scale. † measured at another effort setting of the same model (the publisher did not test the one compared here).

Price, speed and latency

List prices blended at three input tokens per output token. Per-task prices use the token counts on the methodology page; suite cost is what Artificial Analysis paid to run its whole Intelligence Index.

MeasureMuse Spark 1.1Muse Spark 1.2 (xhigh)
Blended price per 1M tokens$2.00$2.00
Input / output per 1M$1.25 / $4.25$1.25 / $4.25
Cached input per 1M$0.150$0.150
Price per task: chat reply$0.0018$0.0018
Price per task: summarize a 30-page report$0.018$0.018
Price per task: code edit$0.014$0.014
Price per task: agentic coding session$0.092$0.092
Price per task: structured extraction$0.0034$0.0034
Cost to run a full benchmark suite–$0.975
Output speed200 tok/s307 tok/s
Time to first answer1.5 s17.6 s
Time to a complete answer–19.2 s

Each model is compared at its best-ranked reasoning-effort setting. Muse Spark 1.1 scores 62.5 and takes 2 s to its first answer token; Muse Spark 1.1 (xhigh) scores 61.6 in 18 s. Muse Spark 1.2 (xhigh) scores 63.2 and takes 18 s to its first answer token. Higher effort means more thinking before the answer: better scores, longer waits and more tokens billed. How reasoning effort changes scores and cost.

Every benchmark

104 benchmarks, grouped by category; each name links to the board and its leaders. † measured at another effort setting of the same model, shown for reference and not counted as a win.

MetricPublisherMuse Spark 1.1Muse Spark 1.2 (xhigh)
BenchLeader IndexBenchLeader62.563.2
Agents & tools scoreBenchLeader51.855.4
Coding scoreBenchLeader66.355.1
Human preference scoreBenchLeader65.068.6
Instruction following scoreBenchLeader71.5–
Knowledge scoreBenchLeader69.266.7
Maths scoreBenchLeader61.367.3
Multimodal scoreBenchLeader63.665.2
Reasoning scoreBenchLeader67.969.3
Composite scoreBenchLeader–73.8
Long context scoreBenchLeader–64.5
Blended price $/MBenchLeader$2.00$2.00
Output speedBenchLeader200 tok/s307 tok/s
Time to first answerBenchLeader1.5 s17.6 s
Context windowBenchLeader1,048,576 tokens1,048,576 tokens
Reasoning
SimpleBenchSimpleBench–74.5%†
Arena Hard PromptsArena (LMArena)15101512
LiveBench 2.0 ReasoningLiveBench–90.0%†
LiveBench 1.0 ReasoningLiveBench87.7%†90.0%
GPQA Diamond (AA)Artificial Analysis89.8%†90.4%
Humanity's Last Exam (AA)Artificial Analysis46.2%†45.5%
GPQA Diamond (Vals)Vals AI91.2%†–
CritPtArtificial Analysis15.1%†17.7%
Chess PuzzlesEpoch AI Benchmarking Hub34.0%†27.0%
LMCAEpoch AI Benchmarking Hub49.9%†48.4%
DTBenchEpoch AI Benchmarking Hub94.4%†94.7%
Coding
SciCodeSciCode58.8%57.4%
WeirdMLWeirdML–60.3%
Arena CodingArena (LMArena)15331541
Arena WebDevArena (LMArena)15431533
LiveBench 2.0 CodingLiveBench–78.2%†
LiveBench 1.0 CodingLiveBench77.2%†77.5%
LiveCodeBenchVals AI85.9%†–
IOIVals AI–21.8%
SWE-bench (Vals)Vals AI82.0%†86.6%
SWE-Bench ProScale AI SEAL61.5%–
Code MigrationVals AI31.1%†29.9%
ProgramBenchVals AI0.0%†0.5%
Vibe Code Bench v1.1Vals AI72.2%†79.1%
SWE-Bench Pro (private)Scale AI SEAL61.5%–
Show all 104 benchmarks
MetricPublisherMuse Spark 1.1Muse Spark 1.2 (xhigh)
Coding
SWE Atlas: Codebase QnAScale AI SEAL42.2%†–
SWE Atlas: Test WritingScale AI SEAL41.5%†–
Surface Evolver BenchEpoch AI Benchmarking Hub52.5%†–
DeepSWE v1.1Epoch AI Benchmarking Hub53.3%54.9%
FrontierSWEEpoch AI Benchmarking Hub–12.0%
Terminal-Bench 4.0 (AA)Artificial Analysis6.1%†7.1%
Terminal-Bench 2.1 (AA)Artificial Analysis77.9%†80.2%
Startrise FrontendStartrise–70.3%†
Agents & tools
APEX-AgentsMercor31.8%45.4%†
Arena AgentArena (LMArena)-6.9-7
LiveBench Long AgenticLiveBench–21.3%†
LiveBench BrowserLiveBench–36.0%†
LiveBench Agentic CodingLiveBench58.5%†57.6%
Terminal-Bench 2.1 (Vals)Vals AI69.3%†69.7%
MCP AtlasScale AI SEAL88.1%–
τ³-bench Bankingτ²-bench40.5%†–
τ³-Banking (AA)Artificial Analysis31.8%†34.9%
CyberBenchVals AI–69.5%
Finance Agent v2Vals AI57.2%†60.6%
Harvey's Legal Agent Benchmark v1Vals AI20.0%†25.4%
Legal Research BenchVals AI38.0%†43.8%
SkillsBenchVals AI59.2%†53.0%
SREBenchVals AI–0.0%
Tax Agent BenchVals AI–32.8%
Terminal-Bench 4.0 (Vals)Vals AI–6.1%
Time Horizon Index: KSPVals AI1.7%†–
GBAEvalEpoch AI Benchmarking Hub7.9%–
Vending-Bench 2Andon Labs6520.5–
AutomationBenchArtificial Analysis38.9%†40.6%
GDP.pdf (AA)Artificial Analysis14.4%†17.4%
EnterpriseOps-GymArtificial Analysis–47.3%
AA-Briefcase v1.1Artificial Analysis–1327
Maths
ProofBenchVals AI39.0%43.0%†
LiveBench MathematicsLiveBench87.1%†91.2%
Arena MathsArena (LMArena)14931491
Knowledge
SimpleQA VerifiedEpoch AI Benchmarking Hub57.8%60.3%
LiveBench 2.0 Data AnalysisLiveBench–76.5%†
LiveBench ResearchLiveBench–67.7%†
LiveBench 1.0 Data AnalysisLiveBench72.5%†76.5%
AA-OmniscienceArtificial Analysis28.1†27.2
MMLU-ProVals AI88.7%†88.3%
LegalBenchVals AI85.0%†85.3%
CorpFinVals AI71.3%†70.9%
TaxEvalVals AI79.7%†80.4%
PRBench FinanceScale AI SEAL55.0%–
PRBench LegalScale AI SEAL57.0%–
MultiNRCScale AI SEAL65.6%–
BioMysteryBenchVals AI–64.8%
Excel Modeling BenchmarkVals AI56.4%†57.0%
MedCodeVals AI–49.4%
MedScribeVals AI88.9%†90.1%
Public Benefits Bench v1.1Vals AI61.3%†68.5%
GDP.pdfEpoch AI Benchmarking Hub15.0%16.0%
AA-Omniscience: accuracyArtificial Analysis52.0%†45.4%
AA-Omniscience: non-hallucinationArtificial Analysis50.0%†66.7%
Instruction following
LiveBench LanguageLiveBench74.3%†78.6%
LiveBench Instruction FollowingLiveBench69.6%†74.3%
MultiChallengeScale AI SEAL75.3%–
Arena Instruction FollowingArena (LMArena)14721472
Human preference
Arena TextArena (LMArena)14901491
EQ-Bench 4EQ-Bench1252–
Arena Creative WritingArena (LMArena)14431439
Arena Multi-turnArena (LMArena)14961511
Arena Longer QueriesArena (LMArena)14801488
Multimodal
Arena VisionArena (LMArena)12931304
LiveBench DesignLiveBench–67.1%†
MMMU-Pro (Vals)Vals AI86.6%†86.1%
MortgageTaxVals AI66.1%†65.4%
SAGEVals AI45.6%†47.7%
VTBScale AI SEAL44.8%–
Long context
AA-LCRArtificial Analysis77.7%†79.0%
Arena DocumentArena (LMArena)1480–
MLCRArtificial Analysis7.2%†7.2%
Composite
Epoch Capabilities IndexEpoch AI Benchmarking Hub154.1155.1†
LiveBench 2.0LiveBench–62.4%†
LiveBench 1.0LiveBench75.3%†78.0%
AA Intelligence Index v4.3.2Artificial Analysis33.7†39.6
Vals Multimodal IndexVals AI66.7%†69.8%†
Safety & honesty
FORTRESSScale AI SEAL12.4%–

Data as of 2026-10-25. Best configuration of each model; every score links to its source on the model pages.

What changed

  • Muse Spark 1.1: 1 score revision.
  • Muse Spark 1.2 (xhigh): 1 new result (AA-Briefcase v1.1).
  • Muse Spark 1.2 (xhigh): 1 new result (Harvey's Legal Agent Benchmark v1).
  • Muse Spark 1.1: BenchLeader Index 63.1 → 62.5.
  • Muse Spark 1.2 (xhigh): 2 score revisions.
  • Muse Spark 1.2 (xhigh): BenchLeader Index 62.6 → 63.3.
  • Muse Spark 1.2 (xhigh): 4 new results (AutomationBench, GDP.pdf (AA), MLCR…).
  • Muse Spark 1.2 (xhigh): 3 new results (Chess Puzzles, ProgramBench, SREBench); 1 score revision.
  • Muse Spark 1.1: 1 score revision.
  • Muse Spark 1.2 (xhigh): 3 score revisions.

Rankings: Best LLM overall · Best LLM for coding · Best LLM for agents · Best cheap LLMs

Muse Spark 1.1 vs Muse Spark 1.2: questions

Is Muse Spark 1.1 better than Muse Spark 1.2?
Muse Spark 1.1 and Muse Spark 1.2 (xhigh) are neck and neck on quality (62.5 vs 63.2, within the ±7.2 margin). The BenchLeader Index combines every independent quality benchmark; the two are within the margin of error as of 2026-10-25, so check the category scores for your use.
Which is cheaper, Muse Spark 1.1 or Muse Spark 1.2, and will that change?
They cost the same: $2.00 per million tokens blended. No price change has been announced for either.
Is Muse Spark 1.1 better than Muse Spark 1.2 for coding?
Muse Spark 1.1 scores higher in coding (66 vs 55 on the category index, where 50 is average). On Arena WebDev, the top shared coding benchmark, it is 1543 for Muse Spark 1.1 against 1533 for Muse Spark 1.2.
Is Muse Spark 1.1 better than Muse Spark 1.2 for agents and tool use?
Muse Spark 1.2 scores higher in agentic tasks (55 vs 52 on the category index, where 50 is average). On Arena Agent, the top shared agents & tools benchmark, it is -6.9 for Muse Spark 1.1 against -7 for Muse Spark 1.2.
Which is better for long documents?
Both accept 1,048,576 tokens of context.
Which has lower latency?
Muse Spark 1.1 answers first: 1.5 s against 17.6 s to the first answer token, as Artificial Analysis measures it. Reasoning effort drives most of this: a higher setting thinks longer before it answers.
Which is faster, Muse Spark 1.1 or Muse Spark 1.2?
Muse Spark 1.2 streams faster: 307 against 200 output tokens per second.
Do independent benchmarks agree on Muse Spark 1.1 vs Muse Spark 1.2?
Across the 14 benchmarks both have, Muse Spark 1.1 leads 5, Muse Spark 1.2 leads 9. Every overall index that rates both puts Muse Spark 1.2 ahead.
When was Muse Spark 1.2 released?
Muse Spark 1.2 was released on 5 Aug 2026.