Enter opens the highlighted result. Esc closes.

Claude Haiku 4.5 vs Muse Spark 1.2

Muse Spark 1.2 leads Claude Haiku 4.5 on coding, 55.7 to 35.6 on the Vibe Coding Index, at about the same output price ($4.25 vs $5.00 per million). Updated October 22, 2026; 72 benchmarks have a result for both.

Replace Claude Haiku 4.5 Replace Muse Spark 1.2
Claude Haiku 4.5Anthropic
Coding
35.6
Overall
41.0
Output $/M
$5.00
1
Muse Spark 1.2Meta
Coding
55.7
Overall
60.4
Output $/M
$4.25
71
vs boards won of 72 shared

Claude Haiku 4.5 or Muse Spark 1.2: which to pick

Pick Muse Spark 1.2 for coding: it scores 55.7 on the Vibe Coding Index against 35.6 for Claude Haiku 4.5.

Pick Claude Haiku 4.5 for a lower cost per solved task ($0.56 against $0.64) and work measured by LiveBench browser, where it leads.

Model AClaude Haiku 4.5Anthropic35.6/1004 independent benchmarks$0.56 per solved taskModel BMuse Spark 1.2Meta55.7/1005 independent benchmarks$0.64 per solved task
Shared results
72
Claude Haiku 4.5 only
23
Muse Spark 1.2 only
25
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
Claude Haiku 4.5 35.6 from 4 benchmarks, Muse Spark 1.2 55.7 from 5 benchmarks.
Widest category
Coding, 25.3 points apart on average over 13 shared benchmarks.
Widest single gap
Vibe Code Bench (Vals AI): Claude Haiku 4.5 11.4% against Muse Spark 1.2 79.1%, 67.7 points apart on a 0 to 100 scale.
Output price
Claude Haiku 4.5 $5.00 against Muse Spark 1.2 $4.25 per million output tokens, 1.2x apart.
Context window
Claude Haiku 4.5 takes 200K tokens, Muse Spark 1.2 1.05M.

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
CategoryClaude Haiku 4.5BarsMuse Spark 1.2Ties
Coding013
Intelligence02
Agentic work17
Reasoning06
Math02
Knowledge011
Web design09
General021
All shared boards171
CodingLike-for-likeSix of these boards feed the Vibe Coding Index013
Coding benchmarks for Claude Haiku 4.5 and Muse Spark 1.2
BenchmarkClaude Haiku 4.5Muse Spark 1.2Leader
SWE-bench Verified66.6%Not measuredOnly Claude Haiku 4.5 measured
SWE-bench Multilingual64.7%Not measuredOnly Claude Haiku 4.5 measured
LiveBench coding48.1%78.2%Muse Spark 1.2
LMArena WebDev13301534Muse Spark 1.2
AA Coding Index43.972.2Muse Spark 1.2
LiveCodeBench (Vals AI)41.2%Not measuredOnly Claude Haiku 4.5 measured
SWE-bench Verified (Vals AI)66.6%86.6%Muse Spark 1.2
Terminal-Bench 4 (Vals AI)Not measured6.1%Only Muse Spark 1.2 measured
Vibe Code Bench (Vals AI)11.4%79.1%Muse Spark 1.2
Code migration (Vals AI)7.5%29.9%Muse Spark 1.2
ProgramBench (Vals AI)0.0%0.5%Muse Spark 1.2
IOI (Vals AI)Not measured21.8%Only Muse Spark 1.2 measured
Terminal-Bench 2.1 (Vals AI)43.8%69.7%Muse Spark 1.2
SkillsBench (Vals AI)Not measured53.0%Only Muse Spark 1.2 measured
CyberBench (Vals AI)Not measured69.5%Only Muse Spark 1.2 measured
SciCodeNot measured57.4%Only Muse Spark 1.2 measured
Terminal-Bench35.5%Not measuredOnly Claude Haiku 4.5 measured
FrontierSWENot measured12.0%Only Muse Spark 1.2 measured
WeirdML45.4%60.3%Muse Spark 1.2
LMArena Text: coding14541505Muse Spark 1.2
LMArena Text: software and IT14391505Muse Spark 1.2
LMArena WebDev: React13261526Muse Spark 1.2
LMArena WebDev: HTML13181543Muse Spark 1.2
Neither model has a result on the other 16 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
Intelligence benchmarks for Claude Haiku 4.5 and Muse Spark 1.2
BenchmarkClaude Haiku 4.5Muse Spark 1.2Leader
AA Intelligence Index16.939.6Muse Spark 1.2
Epoch Capabilities Index142.2155.1Muse Spark 1.2
Agentic workLike-for-likePublic boards, shown as published, kept out of the index17
Agentic work benchmarks for Claude Haiku 4.5 and Muse Spark 1.2
BenchmarkClaude Haiku 4.5Muse Spark 1.2Leader
AA Agentic Index8.043.2Muse Spark 1.2
Design Arena: agents building web appsNot measured1221Only Muse Spark 1.2 measured
Design Arena: agents building full-stack appsNot measured1210Only Muse Spark 1.2 measured
Design Arena: agents building mobile appsNot measured1181Only Muse Spark 1.2 measured
Design Arena: agents building Android appsNot measured1222Only Muse Spark 1.2 measured
Design Arena: agents building gamesNot measured1179Only Muse Spark 1.2 measured
Design Arena: HTML slidesNot measured1085Only Muse Spark 1.2 measured
Design Arena: PowerPoint slidesNot measured1172Only Muse Spark 1.2 measured
LiveBench research42.2%67.7%Muse Spark 1.2
LiveBench browser40.0%36.0%Claude Haiku 4.5
LiveBench long agentic11.1%21.3%Muse Spark 1.2
SREBench (Vals AI)Not measured0.0%Only Muse Spark 1.2 measured
Finance Agent (Vals AI)31.0%60.6%Muse Spark 1.2
Tax Agent Bench (Vals AI)5.9%32.8%Muse Spark 1.2
Legal Research Bench (Vals AI)10.6%43.8%Muse Spark 1.2
Excel Modeling Benchmark (Vals AI)22.7%57.0%Muse Spark 1.2
BioMysteryBench (Vals AI)Not measured64.8%Only Muse Spark 1.2 measured
APEX-AgentsNot measured45.4%Only Muse Spark 1.2 measured
BALROG31.2%Not measuredOnly Claude Haiku 4.5 measured
DeepResearch Bench45.5%Not measuredOnly Claude Haiku 4.5 measured
Vending-Bench 2$459Not measuredOnly Claude Haiku 4.5 measured
Berkeley Function Calling Leaderboard (BFCL)68.7%Not measuredOnly Claude Haiku 4.5 measured
Neither model has a result on the other 19 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index06
Reasoning benchmarks for Claude Haiku 4.5 and Muse Spark 1.2
BenchmarkClaude Haiku 4.5Muse Spark 1.2Leader
GPQA Diamond72.2%Not measuredOnly Claude Haiku 4.5 measured
LiveBench reasoning61.7%90.0%Muse Spark 1.2
MMMU Pro (Vals AI)46.1%86.1%Muse Spark 1.2
GPQA Diamond (Artificial Analysis run)Not measured90.4%Only Muse Spark 1.2 measured
Humanity's Last Exam (Artificial Analysis run)Not measured45.5%Only Muse Spark 1.2 measured
GPQA Diamond (Epoch AI run)71.2%Not measuredOnly Claude Haiku 4.5 measured
ARC-AGI47.7%Not measuredOnly Claude Haiku 4.5 measured
ARC-AGI-24.0%Not measuredOnly Claude Haiku 4.5 measured
SimpleBenchNot measured74.5%Only Muse Spark 1.2 measured
Chess puzzles8.0%27.0%Muse Spark 1.2
LMArena Text: hard prompts14211489Muse Spark 1.2
LMArena Text: expert14431475Muse Spark 1.2
LMArena Text: hard prompts in English14351487Muse Spark 1.2
LMArena Vision: diagramsNot measured1317Only Muse Spark 1.2 measured
Neither model has a result on the other 8 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index02
Math benchmarks for Claude Haiku 4.5 and Muse Spark 1.2
BenchmarkClaude Haiku 4.5Muse Spark 1.2Leader
AIME82.7%Not measuredOnly Claude Haiku 4.5 measured
MGSM (Vals AI)92.2%Not measuredOnly Claude Haiku 4.5 measured
OTIS Mock AIME66.7%Not measuredOnly Claude Haiku 4.5 measured
ProofBenchNot measured43.0%Only Muse Spark 1.2 measured
MATH level 596.4%Not measuredOnly Claude Haiku 4.5 measured
LMArena Text: math13981484Muse Spark 1.2
LMArena Text: mathematical work14301480Muse Spark 1.2
Neither model has a result on the other 6 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index011
Knowledge benchmarks for Claude Haiku 4.5 and Muse Spark 1.2
BenchmarkClaude Haiku 4.5Muse Spark 1.2Leader
MMLU Pro78.7%88.3%Muse Spark 1.2
LegalBench (Vals AI)81.2%85.3%Muse Spark 1.2
CaseLaw (Vals AI)56.5%Not measuredOnly Claude Haiku 4.5 measured
CorpFin (Vals AI)60.6%70.9%Muse Spark 1.2
TaxEval (Vals AI)67.5%80.4%Muse Spark 1.2
MortgageTax (Vals AI)62.2%65.4%Muse Spark 1.2
MedQA (Vals AI)79.6%Not measuredOnly Claude Haiku 4.5 measured
MedCode (Vals AI)32.7%49.4%Muse Spark 1.2
SimpleQA Verified13.2%60.3%Muse Spark 1.2
LMArena Text: science14061500Muse Spark 1.2
LMArena Text: medicine and healthcare13941496Muse Spark 1.2
LMArena Text: legal and government14001517Muse Spark 1.2
LMArena Text: business and finance14001484Muse Spark 1.2
Neither model has a result on the other 5 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index09
Web design benchmarks for Claude Haiku 4.5 and Muse Spark 1.2
BenchmarkClaude Haiku 4.5Muse Spark 1.2Leader
Design Arena: websites11221312Muse Spark 1.2
Design Arena: UI components10971298Muse Spark 1.2
Design Arena: game dev10971301Muse Spark 1.2
Design Arena: data viz11241334Muse Spark 1.2
Design Arena: 3D10711301Muse Spark 1.2
Design Arena: all code categories11141312Muse Spark 1.2
Design Arena: SVG10251284Muse Spark 1.2
Design Arena: ASCII art11401284Muse Spark 1.2
LiveBench design59.9%67.1%Muse Spark 1.2
GeneralLike-for-likePublic boards, shown as published, kept out of the index021
General benchmarks for Claude Haiku 4.5 and Muse Spark 1.2
BenchmarkClaude Haiku 4.5Muse Spark 1.2Leader
LiveBench data analysis59.3%76.5%Muse Spark 1.2
Public Benefits Bench (Vals AI)54.3%68.5%Muse Spark 1.2
SAGE (Vals AI)31.8%47.7%Muse Spark 1.2
MedScribe (Vals AI)85.2%90.1%Muse Spark 1.2
Vals Multimodal Index42.9%69.8%Muse Spark 1.2
Long context reasoning (AA-LCR)Not measured79.0%Only Muse Spark 1.2 measured
DTBench73.6%94.7%Muse Spark 1.2
LMCA30.9%48.4%Muse Spark 1.2
ExploitBench13.7%Not measuredOnly Claude Haiku 4.5 measured
LMArena Text13971482Muse Spark 1.2
LMArena Text: instruction following14141462Muse Spark 1.2
LMArena Text: creative writing13711436Muse Spark 1.2
LMArena VisionNot measured1304Only Muse Spark 1.2 measured
LMArena Text: English14141483Muse Spark 1.2
LMArena Text: non-English13771476Muse Spark 1.2
LMArena Text: Chinese14181509Muse Spark 1.2
LMArena Text: German1374Not measuredOnly Claude Haiku 4.5 measured
LMArena Text: Japanese1338Not measuredOnly Claude Haiku 4.5 measured
LMArena Text: Korean13461462Muse Spark 1.2
LMArena Text: French14111495Muse Spark 1.2
LMArena Text: Spanish14211484Muse Spark 1.2
LMArena Text: Russian13811483Muse Spark 1.2
LMArena Text: Polish1343Not measuredOnly Claude Haiku 4.5 measured
LMArena Text: multi-turn14091502Muse Spark 1.2
LMArena Text: longer queries14271473Muse Spark 1.2
LMArena Text: writing and language13861453Muse Spark 1.2
LMArena Text: entertainment and media13691443Muse Spark 1.2
LMArena Vision: EnglishNot measured1291Only Muse Spark 1.2 measured
LMArena Vision: reading textNot measured1313Only Muse Spark 1.2 measured
LMArena Document1420Not measuredOnly Claude Haiku 4.5 measured
Neither model has a result on the other 16 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Claude Haiku 4.5 and Muse Spark 1.2
BenchmarkClaude Haiku 4.5Muse Spark 1.2Leader
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

Claude Haiku 4.5$0.0035Fits in one request
Muse Spark 1.2$0.0034Fits in one request

Muse Spark 1.2 costs less

Repository review

50K input + 3K output tokens

Claude Haiku 4.5$0.065Fits in one request
Muse Spark 1.2$0.075Fits in one request

Claude Haiku 4.5 costs less

Agent loop

200K input + 10K output tokens

Claude Haiku 4.5$0.250Exceeds the context window
Muse Spark 1.2$0.292Fits in one request

Claude Haiku 4.5 costs less

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 Claude Haiku 4.5 and Muse Spark 1.2
SpecificationAnthropic Claude Haiku 4.5Meta Muse Spark 1.2
MakerAnthropicMeta
WeightsNo public weights listedNo public weights listed
ReleasedOctober 15, 2025August 5, 2026
Context window200K1.05M
Input $/Mtok$1.00$1.25
Output $/Mtok$5.00$4.25
Coding index35.655.7
Independent measurements behind it45
Cost per solved task$0.56$0.64
SourceOpenRouter listingOpenRouter listing

Questions

Is Claude Haiku 4.5 better than Muse Spark 1.2 for coding?

Pick Muse Spark 1.2 for coding: it scores 55.7 on the Vibe Coding Index against 35.6 for Claude Haiku 4.5. Pick Claude Haiku 4.5 for a lower cost per solved task ($0.56 against $0.64) and work measured by LiveBench browser, where it leads.

Which is cheaper, Claude Haiku 4.5 or Muse Spark 1.2?

Claude Haiku 4.5 costs $1.00 per million input tokens and $5.00 per million output tokens; Muse Spark 1.2 costs $1.25 and $4.25. Per solved SWE-bench Verified task, Claude Haiku 4.5 costs $0.56 and Muse Spark 1.2 $0.64.

Which has the bigger context window, Claude Haiku 4.5 or Muse Spark 1.2?

Muse Spark 1.2, with 1.05M tokens against 200K.

How many benchmarks is the Claude Haiku 4.5 vs Muse Spark 1.2 comparison based on?

72 benchmarks have a result for both models as of October 22, 2026. Claude Haiku 4.5's coding index rests on 4 independent benchmarks and Muse Spark 1.2's on 5. 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/claude-haiku-4-5-vs-meta-muse-spark-1-2" title="Claude Haiku 4.5 vs Muse Spark 1.2, 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/claude-haiku-4-5-vs-meta-muse-spark-1-2">LLMBenchmark.ai, Claude Haiku 4.5 vs Muse Spark 1.2</a></p>