Enter opens the highlighted result. Esc closes.

Muse Spark 1.2 vs Muse Spark 1.3

Muse Spark 1.3 leads Muse Spark 1.2 on coding, 57.4 to 55.7 on the Vibe Coding Index, at about the same output price ($4.25 vs $4.25 per million). Updated October 22, 2026; 76 benchmarks have a result for both.

Replace Muse Spark 1.2 Replace Muse Spark 1.3
Muse Spark 1.2Meta
Coding
55.7
Overall
60.4
Output $/M
$4.25
9
Muse Spark 1.3Meta
Coding
57.4
Overall
68.6
Output $/M
$4.25
66
vs boards won of 76 shared, 1 tied

Muse Spark 1.2 or Muse Spark 1.3: which to pick

Muse Spark 1.2 and Muse Spark 1.3 are level for coding, 55.7 against 57.4 on the Vibe Coding Index, so price and context decide.

Pick Muse Spark 1.2 for work measured by Code migration (Vals AI) or CyberBench (Vals AI), where it leads. Pick Muse Spark 1.3 for work measured by LiveBench coding or LMArena WebDev, where it leads.

Model AMuse Spark 1.2Meta55.7/1005 independent benchmarks$0.64 per solved taskModel BMuse Spark 1.3Meta57.4/1004 independent benchmarksNo cost per solved task
Shared results
76
Muse Spark 1.2 only
21
Muse Spark 1.3 only
20
Like-for-like categories
8 of 9

Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.

Which one for your work

Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.

At a glance

Coding index
Muse Spark 1.2 55.7 from 5 benchmarks, Muse Spark 1.3 57.4 from 4 benchmarks.
Widest category
Agentic work, 10.3 points apart on average over 13 shared benchmarks.
Widest single gap
LiveBench browser: Muse Spark 1.2 36.0% against Muse Spark 1.3 80.0%, 44.0 points apart on a 0 to 100 scale.
Output price
Both cost $4.25 per million output tokens.
Context window
Both take 1.05M tokens.

Benchmark by benchmark

First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.

Boards won, where both have a result
CategoryMuse Spark 1.2BarsMuse Spark 1.3Ties
Coding2131 tie
Intelligence02
Agentic work211
Reasoning17
Math03
Knowledge22
Web design18
General120
All shared boards9661 tie
CodingLike-for-likeSix of these boards feed the Vibe Coding Index213
Coding benchmarks for Muse Spark 1.2 and Muse Spark 1.3
BenchmarkMuse Spark 1.2Muse Spark 1.3Leader
LiveBench coding78.2%90.2%Muse Spark 1.3
LMArena WebDev15341659Muse Spark 1.3
AA Coding Index72.275.8Muse Spark 1.3
SWE-bench Verified (Vals AI)86.6%Not measuredOnly Muse Spark 1.2 measured
Terminal-Bench 4 (Vals AI)6.1%10.6%Muse Spark 1.3
Vibe Code Bench (Vals AI)79.1%82.9%Muse Spark 1.3
Code migration (Vals AI)29.9%27.6%Muse Spark 1.2
ProgramBench (Vals AI)0.5%0.5%Tie
IOI (Vals AI)21.8%43.9%Muse Spark 1.3
Terminal-Bench 2.1 (Vals AI)69.7%72.3%Muse Spark 1.3
SkillsBench (Vals AI)53.0%Not measuredOnly Muse Spark 1.2 measured
CyberBench (Vals AI)69.5%69.4%Muse Spark 1.2
SciCode57.4%59.7%Muse Spark 1.3
FrontierSWE12.0%25.8%Muse Spark 1.3
WeirdML60.3%Not measuredOnly Muse Spark 1.2 measured
ALE-BenchNot measured1688.3Only Muse Spark 1.3 measured
LMArena Text: coding15051510Muse Spark 1.3
LMArena Text: software and IT15051511Muse Spark 1.3
LMArena WebDev: React15261672Muse Spark 1.3
LMArena WebDev: HTML15431653Muse Spark 1.3
LMArena WebDev: from an imageNot measured1640Only Muse Spark 1.3 measured
Neither model has a result on the other 22 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
Intelligence benchmarks for Muse Spark 1.2 and Muse Spark 1.3
BenchmarkMuse Spark 1.2Muse Spark 1.3Leader
AA Intelligence Index39.648.1Muse Spark 1.3
Epoch Capabilities Index155.1157.2Muse Spark 1.3
Agentic workLike-for-likePublic boards, shown as published, kept out of the index211
Agentic work benchmarks for Muse Spark 1.2 and Muse Spark 1.3
BenchmarkMuse Spark 1.2Muse Spark 1.3Leader
AA Agentic Index43.255.5Muse Spark 1.3
Design Arena: agents building web apps12211297Muse Spark 1.3
Design Arena: agents building full-stack apps12101295Muse Spark 1.3
Design Arena: agents building mobile apps11811217Muse Spark 1.3
Design Arena: agents building Android apps1222Not measuredOnly Muse Spark 1.2 measured
Design Arena: agents building Godot gamesNot measured1241Only Muse Spark 1.3 measured
Design Arena: agents building games1178Not measuredOnly Muse Spark 1.2 measured
Design Arena: HTML slides1074Not measuredOnly Muse Spark 1.2 measured
Design Arena: PowerPoint slides11701217Muse Spark 1.3
Design Arena: spreadsheetsNot measured1075Only Muse Spark 1.3 measured
LiveBench research67.7%68.6%Muse Spark 1.3
LiveBench browser36.0%80.0%Muse Spark 1.3
LiveBench long agentic21.3%40.4%Muse Spark 1.3
SREBench (Vals AI)0.0%Not measuredOnly Muse Spark 1.2 measured
Finance Agent (Vals AI)60.6%58.9%Muse Spark 1.2
Tax Agent Bench (Vals AI)32.8%36.0%Muse Spark 1.3
Legal Research Bench (Vals AI)43.8%40.9%Muse Spark 1.2
Excel Modeling Benchmark (Vals AI)57.0%62.7%Muse Spark 1.3
BioMysteryBench (Vals AI)64.8%Not measuredOnly Muse Spark 1.2 measured
Terminal-Bench-Science (Vals AI)Not measured4.3%Only Muse Spark 1.3 measured
APEX-Agents45.4%58.6%Muse Spark 1.3
Neither model has a result on the other 21 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index17
Reasoning benchmarks for Muse Spark 1.2 and Muse Spark 1.3
BenchmarkMuse Spark 1.2Muse Spark 1.3Leader
LiveBench reasoning90.0%89.7%Muse Spark 1.2
MMMU Pro (Vals AI)86.1%Not measuredOnly Muse Spark 1.2 measured
GPQA Diamond (Artificial Analysis run)90.4%94.1%Muse Spark 1.3
Humanity's Last Exam (Artificial Analysis run)45.5%48.7%Muse Spark 1.3
SimpleBench74.5%Not measuredOnly Muse Spark 1.2 measured
Chess puzzles27.0%37.0%Muse Spark 1.3
Mystery game puzzlesNot measured25.0%Only Muse Spark 1.3 measured
LMArena Text: hard prompts14891503Muse Spark 1.3
LMArena Text: expert14751519Muse Spark 1.3
LMArena Text: hard prompts in English14871500Muse Spark 1.3
LMArena Vision: diagrams13171330Muse Spark 1.3
LMArena Vision: homeworkNot measured1296Only Muse Spark 1.3 measured
Humanity's Last Exam (Scale run)Not measured25.4%Only Muse Spark 1.3 measured
Neither model has a result on the other 10 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index03
Math benchmarks for Muse Spark 1.2 and Muse Spark 1.3
BenchmarkMuse Spark 1.2Muse Spark 1.3Leader
OTIS Mock AIMENot measured99.2%Only Muse Spark 1.3 measured
FrontierMath, tiers 1 to 3Not measured74.4%Only Muse Spark 1.3 measured
FrontierMath, tier 4Not measured46.3%Only Muse Spark 1.3 measured
ProofBench43.0%58.0%Muse Spark 1.3
LMArena Text: math14841497Muse Spark 1.3
LMArena Text: mathematical work14801491Muse Spark 1.3
MathArenaNot measured43.0%Only Muse Spark 1.3 measured
Neither model has a result on the other 7 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index22
Knowledge benchmarks for Muse Spark 1.2 and Muse Spark 1.3
BenchmarkMuse Spark 1.2Muse Spark 1.3Leader
MMLU Pro88.3%Not measuredOnly Muse Spark 1.2 measured
LegalBench (Vals AI)85.3%Not measuredOnly Muse Spark 1.2 measured
CorpFin (Vals AI)70.9%Not measuredOnly Muse Spark 1.2 measured
TaxEval (Vals AI)80.4%Not measuredOnly Muse Spark 1.2 measured
MortgageTax (Vals AI)65.4%Not measuredOnly Muse Spark 1.2 measured
MedCode (Vals AI)49.4%Not measuredOnly Muse Spark 1.2 measured
SimpleQA Verified60.3%Not measuredOnly Muse Spark 1.2 measured
LMArena Text: science15001507Muse Spark 1.3
LMArena Text: medicine and healthcare14961493Muse Spark 1.2
LMArena Text: legal and government15171499Muse Spark 1.2
LMArena Text: business and finance14841490Muse Spark 1.3
PRBench FinanceNot measured59.5%Only Muse Spark 1.3 measured
PRBench LegalNot measured61.6%Only Muse Spark 1.3 measured
Neither model has a result on the other 7 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index18
Web design benchmarks for Muse Spark 1.2 and Muse Spark 1.3
BenchmarkMuse Spark 1.2Muse Spark 1.3Leader
Design Arena: websites13121356Muse Spark 1.3
Design Arena: UI components12991350Muse Spark 1.3
Design Arena: game dev13001328Muse Spark 1.3
Design Arena: data viz13341344Muse Spark 1.3
Design Arena: 3D13001358Muse Spark 1.3
Design Arena: all code categories13121352Muse Spark 1.3
Design Arena: SVG12831301Muse Spark 1.3
Design Arena: ASCII art12841283Muse Spark 1.2
LiveBench design67.1%74.8%Muse Spark 1.3
GeneralLike-for-likePublic boards, shown as published, kept out of the index120
General benchmarks for Muse Spark 1.2 and Muse Spark 1.3
BenchmarkMuse Spark 1.2Muse Spark 1.3Leader
LiveBench data analysis76.5%79.6%Muse Spark 1.3
Public Benefits Bench (Vals AI)68.5%Not measuredOnly Muse Spark 1.2 measured
SAGE (Vals AI)47.7%Not measuredOnly Muse Spark 1.2 measured
MedScribe (Vals AI)90.1%Not measuredOnly Muse Spark 1.2 measured
Vals Multimodal Index69.8%Not measuredOnly Muse Spark 1.2 measured
Long context reasoning (AA-LCR)79.0%83.0%Muse Spark 1.3
DTBench94.7%96.5%Muse Spark 1.3
LMCA48.4%53.9%Muse Spark 1.3
LMArena Text14821490Muse Spark 1.3
LMArena Text: instruction following14621479Muse Spark 1.3
LMArena Text: creative writing14361462Muse Spark 1.3
LMArena Vision13041308Muse Spark 1.3
LMArena Text: English14831495Muse Spark 1.3
LMArena Text: non-English14761481Muse Spark 1.3
LMArena Text: Chinese15091532Muse Spark 1.3
LMArena Text: GermanNot measured1513Only Muse Spark 1.3 measured
LMArena Text: JapaneseNot measured1470Only Muse Spark 1.3 measured
LMArena Text: Korean14621490Muse Spark 1.3
LMArena Text: French14951511Muse Spark 1.3
LMArena Text: Spanish14841496Muse Spark 1.3
LMArena Text: Russian14831491Muse Spark 1.3
LMArena Text: PolishNot measured1490Only Muse Spark 1.3 measured
LMArena Text: multi-turn15021487Muse Spark 1.2
LMArena Text: longer queries14731488Muse Spark 1.3
LMArena Text: writing and language14531470Muse Spark 1.3
LMArena Text: entertainment and media14431457Muse Spark 1.3
LMArena Vision: English12911298Muse Spark 1.3
LMArena Vision: ChineseNot measured1384Only Muse Spark 1.3 measured
LMArena Vision: reading text13131319Muse Spark 1.3
LMArena Vision: humorNot measured1293Only Muse Spark 1.3 measured
LMArena Vision: creative writingNot measured1296Only Muse Spark 1.3 measured
Neither model has a result on the other 17 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Muse Spark 1.2 and Muse Spark 1.3
BenchmarkMuse Spark 1.2Muse Spark 1.3Leader
Neither model has a result on 5 boards here

What each workload costs

Three fixed token mixes at each model's listed price, so per-token rates become a bill.

Chat turn

1K input + 500 output tokens

Muse Spark 1.2$0.0034Fits in one request
Muse Spark 1.3$0.0034Fits in one request

Same cost

Repository review

50K input + 3K output tokens

Muse Spark 1.2$0.075Fits in one request
Muse Spark 1.3$0.075Fits in one request

Same cost

Agent loop

200K input + 10K output tokens

Muse Spark 1.2$0.292Fits in one request
Muse Spark 1.3$0.292Fits in one request

Same cost

Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.

Specification differences

The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.

Specifications of Muse Spark 1.2 and Muse Spark 1.3
SpecificationMeta Muse Spark 1.2Meta Muse Spark 1.3
MakerMetaMeta
WeightsNo public weights listedNo public weights listed
ReleasedAugust 5, 2026September 2, 2026
Context window1.05M1.05M
Input $/Mtok$1.25$1.25
Output $/Mtok$4.25$4.25
Coding index55.757.4
Independent measurements behind it54
Cost per solved task$0.64Not computable
SourceOpenRouter listingOpenRouter listing

Questions

Is Muse Spark 1.2 better than Muse Spark 1.3 for coding?

Muse Spark 1.2 and Muse Spark 1.3 are level for coding, 55.7 against 57.4 on the Vibe Coding Index, so price and context decide. Pick Muse Spark 1.2 for work measured by Code migration (Vals AI) or CyberBench (Vals AI), where it leads. Pick Muse Spark 1.3 for work measured by LiveBench coding or LMArena WebDev, where it leads.

Which is cheaper, Muse Spark 1.2 or Muse Spark 1.3?

Muse Spark 1.2 costs $1.25 per million input tokens and $4.25 per million output tokens; Muse Spark 1.3 costs $1.25 and $4.25. Cost per solved task is known for Muse Spark 1.2 only.

Which has the bigger context window, Muse Spark 1.2 or Muse Spark 1.3?

Neither: both take 1.05M tokens.

How many benchmarks is the Muse Spark 1.2 vs Muse Spark 1.3 comparison based on?

76 benchmarks have a result for both models as of October 22, 2026. Muse Spark 1.2's coding index rests on 5 independent benchmarks and Muse Spark 1.3's on 4. Where only one model has a result, the page says so instead of naming a winner.

Compare next

Embed this comparison on your site

Free under CC BY 4.0, while the source link stays visible. Preview the frame · Terms and other embeds

<iframe src="https://llmbenchmark.ai/embed/compare/meta-muse-spark-1-2-vs-meta-muse-spark-1-3" title="Muse Spark 1.2 vs Muse Spark 1.3, LLMBenchmark.ai" width="100%" height="360" style="border:0;max-width:100%" loading="lazy"></iframe>
<p style="font:13px/1.4 sans-serif;margin:6px 0 0">Source: <a href="https://llmbenchmark.ai/compare/meta-muse-spark-1-2-vs-meta-muse-spark-1-3">LLMBenchmark.ai, Muse Spark 1.2 vs Muse Spark 1.3</a></p>