BenchLeader
Menu
Best LLMsChoose a modelCoding plansLeaderboardsModelsBenchmarksProvidersReleases4

Muse Spark 1.3 vs Qwen3.6 Max

Full profiles: Muse Spark 1.3 (Meta) and Qwen3.6 Max (Alibaba): every benchmark, price and rank.

Updated · Data from 6 sources

As of 2 Nov 2026, Muse Spark 1.3 and Qwen3.6 Max are neck and neck on quality (61.9 vs 62.4 on the BenchLeader Index, inside the ±4.5 margin), and Muse Spark 1.3 is 1.5× cheaper at $2.00 vs $2.92 per 1M tokens blended. Choose Muse Spark 1.3 for coding (61 vs 56), human preference (69 vs 65), and long context (66 vs 65); choose Qwen3.6 Max for agents & tools (72 vs 66), reasoning (63 vs 61), and maths (65 vs 63).

Key facts (as of 2 Nov 2026)

  • Muse Spark 1.3 scores 61.9 on the BenchLeader Index and Qwen3.6 Max scores 62.4 (BenchLeader, 2 Nov 2026).
  • Muse Spark 1.3 costs $2.00 per 1M tokens blended (list price, 2 Nov 2026).
  • Qwen3.6 Max costs $2.92 per 1M tokens blended (list price, 2 Nov 2026).
  • Of the 13 benchmarks both models have, Muse Spark 1.3 leads 13 (BenchLeader, 2 Nov 2026).
  • On Arena WebDev, Muse Spark 1.3 is rated 1656 and Qwen3.6 Max 1482 (LMArena, as published by 2 Nov 2026).
Muse Spark 1.3 benchmarks max Meta Qwen3.6 Max benchmarks max AlibabaAdd a third model

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 1.361 vs 56 category score; Arena WebDev 1656 vs 1482Medium
Agents & tool useQwen3.6 Max72 vs 66 category scoreMedium
MathsQwen3.6 Max65 vs 63 category score; Arena Maths 1476 vs 1508Medium
ReasoningQwen3.6 Max63 vs 61 category score; Arena Hard Prompts 1482 vs 1516Medium
KnowledgeQwen3.6 Max60 vs 59 category scoreMedium
Long documentsMuse Spark 1.366 vs 65 category score; AA-LCR v1.1 83.0% vs 80.7%Medium
Chat quality (human preference)Muse Spark 1.369 vs 65 on Arena votesMedium
Lowest costMuse Spark 1.3$2.00 vs $2.96 per 1M blended; agentic coding session $0.092 vs $0.111–
High volume on a budgetNeither flagshipA smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs–

Key differences

  • Muse Spark 1.3 (max) and Qwen3.6 Max (max) are neck and neck on quality (61.9 vs 62.4, within the ±4.5 margin).
  • Muse Spark 1.3 (max) is stronger in coding, human preference, long context, multimodal.
  • Qwen3.6 Max (max) is stronger in agents & tools, knowledge, maths, reasoning, instruction following.
  • Muse Spark 1.3 (max) is 1.5× cheaper ($2.00 vs $2.96 per 1M blended).

At a glance

SpecMuse Spark 1.3Qwen3.6 Max
DeveloperMetaAlibaba
Released2 Sep 202620 Apr 2026
Context window (tokens)1,048,576245,800
Compared atMuse Spark 1.3 (max)Qwen3.6 Max (max)
Open weightsNoNo
Price in / out / cached input per 1M$1.25 / $4.25 / $0.150$1.32 / $7.90 / $0.132

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 1.3 (max)Qwen3.6 Max (max)Leader
Reasoning60.662.7Qwen3.6 Max
Coding60.856.3Muse Spark 1.3
Agents & tools65.572.1Qwen3.6 Max
Maths63.164.7Qwen3.6 Max
Knowledge58.560.0Qwen3.6 Max
Instruction following–74.6not comparable
Human preference68.864.6Muse Spark 1.3
Multimodal65.7–not comparable
Long context66.465.2Muse Spark 1.3

On one chart

Each domain on the index's own scale, where 50 is an average model. Click a domain to see the benchmarks behind it. Pick up to 5 models from the list.

2550 avg75ReasoningCodingAgents & toolsMathsKnowledgeLong contextHumanpreferenceMuse Spark 1.3, Reasoning: 60.6Muse Spark 1.3, Coding: 60.8Muse Spark 1.3, Agents & tools: 65.5Muse Spark 1.3, Maths: 63.1Muse Spark 1.3, Knowledge: 58.5Muse Spark 1.3, Long context: 66.4Muse Spark 1.3, Human preference: 68.8Qwen3.6 Max, Reasoning: 62.7Qwen3.6 Max, Coding: 56.3Qwen3.6 Max, Agents & tools: 72.1Qwen3.6 Max, Maths: 64.7Qwen3.6 Max, Knowledge: 60.0Qwen3.6 Max, Long context: 65.2Qwen3.6 Max, Human preference: 64.6

Not drawn, measured for fewer than two of these models: Instruction following, Multimodal.

ModelIndex

Where sources disagree

Across the 13 benchmarks both have, Muse Spark 1.3 leads 13, Qwen3.6 Max leads 0. Every overall index that rates both puts Muse Spark 1.3 ahead.

Overall indexMuse Spark 1.3 (max)Qwen3.6 Max (max)Leader
BenchLeader Index61.962.4neck and neck
Arena Text14951460Muse Spark 1.3

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. Cost, tokens and time per benchmark task are measured: Artificial Analysis’ averages over its whole Intelligence Index run.

MeasureMuse Spark 1.3 (max)Qwen3.6 Max (max)
Blended price per 1M tokens$2.00$2.96
Input / output per 1M$1.25 / $4.25$1.32 / $7.90
Cached input per 1M$0.150$0.132
Price per task: chat reply$0.0018$0.0029
Price per task: summarize a 30-page report$0.018$0.021
Price per task: code edit$0.014$0.020
Price per task: agentic coding session$0.092$0.111
Price per task: structured extraction$0.0034$0.0042
Agentic coding session, input mostly cached$0.043$0.057
Cost per benchmark task (AA Index)$2.39–
Output tokens per task (AA Index)91k–
Of which reasoning55k–
Time per task (AA Index)8 min–
Output speed194 tok/s–
Time to first answer46.3 s–
Time to a complete answer48.9 s–

Each model is compared at its best-ranked reasoning-effort setting. Muse Spark 1.3 (max) scores 61.9 and takes 46 s to its first answer token; Muse Spark 1.3 (xhigh) scores 58.5 in 22 s, Muse Spark 1.3 scores 57.8 in 2 s. Qwen3.6 Max (max) scores 62.4. 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

111 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 1.3 (max)Qwen3.6 Max (max)
BenchLeader IndexBenchLeader61.962.4
Agents & tools scoreBenchLeader65.572.1
Coding scoreBenchLeader60.856.3
Human preference scoreBenchLeader68.864.6
Knowledge scoreBenchLeader58.560.0
Long context scoreBenchLeader66.465.2
Maths scoreBenchLeader63.164.7
Multimodal scoreBenchLeader65.7–
Reasoning scoreBenchLeader60.662.7
Instruction following scoreBenchLeader–74.6
Blended price $/MBenchLeader$2.00$2.96
Output speedBenchLeader194 tok/s–
Time to first answerBenchLeader46.3 s–
Context windowBenchLeader1,048,576 tokens245,800 tokens
Reasoning
GPQA DiamondEpoch AI Benchmarking Hub–87.4%
SimpleBenchSimpleBench–63.0%
Arena Hard PromptsArena (LMArena)15161482
LiveBench 2.0 ReasoningLiveBench89.7%†–
LiveBench 1.0 ReasoningLiveBench89.7%†–
GPQA Diamond (AA)Artificial Analysis93.5%88.8%
Humanity's Last Exam (AA)Artificial Analysis48.7%30.8%
HLE-Diamond (AA)Artificial Analysis33.7%–
CritPtArtificial Analysis24.9%3.7%
CritPt v2 hard subset (AA)Artificial Analysis46.0%–
MysteryMechanismVals AI36.0%–
Chess PuzzlesEpoch AI Benchmarking Hub37.0%–
Mystery Game PuzzlesEpoch AI Benchmarking Hub25.0%19.0%†
LMCAEpoch AI Benchmarking Hub53.9%†42.5%
DTBenchEpoch AI Benchmarking Hub96.5%†87.2%
BTF-3Epoch AI Benchmarking Hub14.1%†–
Coding
SWE-bench Verified (Epoch)Epoch AI Benchmarking Hub–76.7%
SciCodeSciCode58.8%–
Arena CodingArena (LMArena)15341511
Arena WebDevArena (LMArena)16561482
LiveBench 2.0 CodingLiveBench90.2%†–
LiveBench 1.0 CodingLiveBench81.1%†–
IOIVals AI56.6%–
SWE-bench (Vals)Vals AI–72.8%
Code MigrationVals AI47.4%–
Show all 111 benchmarks
MetricPublisherMuse Spark 1.3 (max)Qwen3.6 Max (max)
Coding
ProgramBenchVals AI2.5%–
SWE-bench MultimodalVals AI43.1%–
Vibe Code Bench 1-100Vals AI20.5%–
Vibe Code Bench v1.1Vals AI85.9%–
CursorBenchCursor41.6%–
ALE-BenchEpoch AI Benchmarking Hub1688.3†–
FrontierSWEEpoch AI Benchmarking Hub25.8%–
Terminal-Bench 4.0 (AA)Artificial Analysis33.3%–
Terminal-Bench 2.1 (AA)Artificial Analysis84.3%–
Startrise FrontendStartrise74.7%†–
GameBench 2 OverallDuelLab GameBench 259.7†–
GameBench 2 by reasoning settingDuelLab GameBench 270.9†–
FrontierCode 1.1 (AA)Artificial Analysis40.6%–
FrontierCode 2.0 (AA)Artificial Analysis23.1%–
Agents & tools
Terminal-BenchTerminal-Bench14.6%†–
APEX-AgentsMercor47.6%–
Arena AgentArena (LMArena)2.6–
LiveBench Long AgenticLiveBench25.5%†–
LiveBench BrowserLiveBench30.9%†–
LiveBench Agentic CodingLiveBench64.1%†–
Terminal-Bench HardArtificial Analysis–43.9%
τ²-Bench Telecom (AA)Artificial Analysis–95.9%
Terminal-Bench 2.1 (Vals)Vals AI79.0%–
HiL-BenchScale AI SEAL66.9%†–
τ³-Banking (AA)Artificial Analysis50.5%–
ITBench SRE (AA)Artificial Analysis33.2%–
Analyst Agent (AA)Artificial Analysis37.5%–
CUA-benchVals AI5.8%–
CyberBenchVals AI72.7%–
Finance Agent v2Vals AI60.0%–
Harvey's Legal Agent Benchmark v1Vals AI23.8%–
Legal Research BenchVals AI55.3%–
SREBenchVals AI3.8%–
Tax Agent BenchVals AI42.4%–
Terminal-Bench 2.0 (Vals)Vals AI–51.7%
Terminal-Bench 4.0 (Vals)Vals AI24.8%–
Terminal-Bench ScienceVals AI10.0%–
Vending-Bench 2Andon Labs–4254.2
AutomationBenchArtificial Analysis57.9%–
GDP.pdf (AA)Artificial Analysis26.6%–
Harvey LAB-AA v1.1Harvey8.9%–
AA-Briefcase v1.1Artificial Analysis1581–
Terminal-Bench-Science 0.1 (AA)Artificial Analysis10.9%–
Maths
FrontierMath Tiers 1–3Epoch AI Benchmarking Hub74.0%–
FrontierMath Tier 4Epoch AI Benchmarking Hub46.3%–
OTIS Mock AIMEEpoch AI Benchmarking Hub99.2%†91.1%
ProofBenchVals AI58.0%–
LiveBench MathematicsLiveBench96.0%†–
Arena MathsArena (LMArena)15081476
Knowledge
SimpleQA VerifiedEpoch AI Benchmarking Hub–52.0%
LiveBench 2.0 Data AnalysisLiveBench79.6%†–
LiveBench ResearchLiveBench68.6%†–
LiveBench 1.0 Data AnalysisLiveBench79.6%†–
AA-Omniscience v1.1Artificial Analysis22.4–
CorpFinVals AI–66.5%
PRBench FinanceScale AI SEAL59.5%†–
PRBench LegalScale AI SEAL61.6%†–
CaseLaw v2Vals AI–47.9%
Excel Modeling BenchmarkVals AI67.4%–
DrugDiscoveryBenchScale AI SEAL62.2%†–
GDP.pdfEpoch AI Benchmarking Hub27.6%†–
AA-Omniscience v1.1: accuracyArtificial Analysis38.0%–
AA-Omniscience v1.1: non-hallucinationArtificial Analysis74.8%–
Instruction following
LiveBench LanguageLiveBench82.8%†–
LiveBench Instruction FollowingLiveBench78.0%†–
IFBenchArtificial Analysis–76.6%
Arena Instruction FollowingArena (LMArena)14831451
Human preference
Arena TextArena (LMArena)14951460
Arena Creative WritingArena (LMArena)14641437
Arena Multi-turnArena (LMArena)14931467
Arena Longer QueriesArena (LMArena)14961473
Multimodal
Arena VisionArena (LMArena)1308–
LiveBench DesignLiveBench74.8%†–
MMMU-ProArtificial Analysis82.0%†–
Code2Video-AAArtificial Analysis1019–
AA-3DArtificial Analysis1033–
Long context
AA-LCR v1.1Artificial Analysis83.0%80.7%
Arena DocumentArena (LMArena)1468–
MLCRArtificial Analysis13.9%–
Composite
Epoch Capabilities IndexEpoch AI Benchmarking Hub157.4†149.4
LiveBench 2.0LiveBench63.0%†–
LiveBench 1.0LiveBench81.6%†–
AA Intelligence Index v5.0Artificial Analysis36.4–
AA Intelligence Index v4.3.2Artificial Analysis48.128.4
AA Progress Index v1Artificial Analysis324.5297.1
Vals RSI IndexVals AI19.6%–

Data as of 2026-11-02. Best configuration of each model; every score links to its source on the model pages.

DuelLab GameBench 2, release gb2-20261026-cb69502a12d9 (https://benchmarks.duellab.org/releases/gb2-20261026-cb69502a12d9/), CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Scores mapped to BenchLeader model names; not endorsed by DuelLab.

What changed

  • Muse Spark 1.3 (max): 1 score revision.
  • Muse Spark 1.3 (max): BenchLeader Index 64.5 → 61.8.
  • Muse Spark 1.3 (max): 5 new results (AA Intelligence Index v5.0, AA Progress Index v1, HLE-Diamond (AA)…); 3 score revisions; cost per task moved from $1.31/M to $2.39/M (81% up).
  • Qwen3.6 Max (max): 1 new result (AA Progress Index v1).
  • Muse Spark 1.3 (max): cost per task moved from $3.69/M to $1.31/M (64% down).
  • Muse Spark 1.3 (max): cost per task moved from $1.60/M to $3.69/M (130% up).
  • Muse Spark 1.3 (max): 3 new results (Harvey's Legal Agent Benchmark v1, Vals RSI Index, SWE-bench Multimodal); 1 score revision.
  • Qwen3.6 Max (max): BenchLeader Index 63.3 → 62.8.
  • Muse Spark 1.3 (max): 1 score revision.
  • Muse Spark 1.3 (max): 1 score revision.

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

Muse Spark 1.3 vs Qwen3.6 Max: questions

Is Muse Spark 1.3 better than Qwen3.6 Max?
Muse Spark 1.3 (max) and Qwen3.6 Max (max) are neck and neck on quality (61.9 vs 62.4, within the ±4.5 margin). The BenchLeader Index combines every independent quality benchmark; the two are within the margin of error as of 2 Nov 2026, so check the category scores for your use.
Muse Spark 1.3 vs Qwen3.6 Max: which is cheaper, and will that change?
Muse Spark 1.3 is cheaper today: $2.00 against $2.92 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
Is Muse Spark 1.3 or Qwen3.6 Max better for coding?
Muse Spark 1.3 scores higher in coding (61 vs 56 on the category index, where 50 is average). On Arena WebDev, the top shared coding benchmark, it is 1656 for Muse Spark 1.3 against 1482 for Qwen3.6 Max.
Is Muse Spark 1.3 or Qwen3.6 Max better for agents and tool use?
Qwen3.6 Max scores higher in agentic tasks (72 vs 66 on the category index, where 50 is average).
Which is better for long documents?
Muse Spark 1.3 scores higher on long-context benchmarks (66 vs 65). Muse Spark 1.3 accepts more context: 1,048,576 against 245,800 tokens for Qwen3.6 Max.
More questions (2)
Do independent benchmarks agree on Muse Spark 1.3 vs Qwen3.6 Max?
Across the 13 benchmarks both have, Muse Spark 1.3 leads 13, Qwen3.6 Max leads 0. Every overall index that rates both puts Muse Spark 1.3 ahead.
When was Muse Spark 1.3 released?
Muse Spark 1.3 was released on 2 Sep 2026.