Muse Spark 1.2 vs Muse Wonder
Muse Wonder leads Muse Spark 1.2 on coding, 71.9 to 55.7 on the Vibe Coding Index, and Muse Spark 1.2 costs 4.7x less per output token ($4.25 vs $20.00 per million). Updated October 22, 2026; 50 benchmarks have a result for both.
Muse Spark 1.2 or Muse Wonder: which to pick
Pick Muse Wonder for coding: it scores 71.9 on the Vibe Coding Index against 55.7 for Muse Spark 1.2, though on 3 benchmarks against 5.
Pick Muse Spark 1.2 for output tokens 4.7x cheaper ($4.25 vs $20.00 per million) and work measured by Finance Agent (Vals AI) or LMArena Text: legal and government, where it leads.
- Shared results
- 50
- Muse Spark 1.2 only
- 47
- Muse Wonder only
- 10
- 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.
Coding
Vibe Coding Index, six public coding benchmarks
Muse Wonder
Muse Wonder scores 71.9 against 55.7.
5 and 3 benchmarks behind the scores
Intelligence
AA Intelligence Index
No call
Only Muse Spark 1.2 has a AA Intelligence Index result, 39.6.
Directional only
Agentic work
AA Agentic Index
No call
Only Muse Spark 1.2 has a AA Agentic Index result, 43.2.
Directional only
Knowledge
MMLU Pro
No call
Only Muse Spark 1.2 has a MMLU Pro result, 88.3%.
Directional only
Web design
Design Arena: websites
Muse Wonder
Muse Wonder leads, 1363 against 1312.
Same board, both measured
General
Vals Index
No call
Only Muse Wonder has a Vals Index result, 64.9%.
Directional only
Long documents
Listed context window
No call
Both take 1.05M tokens.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Muse Spark 1.2
Muse Spark 1.2 $0.0034, Muse Wonder $0.014 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
No call
Only Muse Spark 1.2 has the number, $0.64.
Directional only
At a glance
- Coding index
- Muse Spark 1.2 55.7 from 5 benchmarks, Muse Wonder 71.9 from 3 benchmarks.
- Widest category
- Coding, 17.9 points apart on average over 11 shared benchmarks.
- Widest single gap
- Terminal-Bench 4 (Vals AI): Muse Spark 1.2 6.1% against Muse Wonder 49.5%, 57.9 points apart on a 0 to 100 scale.
- Output price
- Muse Spark 1.2 $4.25 against Muse Wonder $20.00 per million output tokens, 4.7x apart.
- 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.
| Category | Muse Spark 1.2 | Bars | Muse Wonder | Ties |
|---|---|---|---|---|
| Coding | 0 | 11 | ||
| Intelligence | 0 | 1 | ||
| Agentic work | 1 | 4 | ||
| Reasoning | 0 | 4 | ||
| Math | 0 | 3 | ||
| Knowledge | 2 | 2 | ||
| Web design | 1 | 5 | ||
| General | 0 | 16 | ||
| All shared boards | 4 | 46 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index011
| Benchmark | Muse Spark 1.2 | Muse Wonder | Leader |
|---|---|---|---|
| LiveBench coding | 78.2% | Not measured | Only Muse Spark 1.2 measured |
| LMArena WebDev | 1534 | 1652 | Muse Wonder |
| AA Coding Index | 72.2 | Not measured | Only Muse Spark 1.2 measured |
| SWE-bench Verified (Vals AI) | 86.6% | Not measured | Only Muse Spark 1.2 measured |
| Terminal-Bench 4 (Vals AI) | 6.1% | 49.5% | Muse Wonder |
| Vibe Code Bench (Vals AI) | 79.1% | 83.1% | Muse Wonder |
| Code migration (Vals AI) | 29.9% | 62.5% | Muse Wonder |
| ProgramBench (Vals AI) | 0.5% | Not measured | Only Muse Spark 1.2 measured |
| IOI (Vals AI) | 21.8% | 66.9% | Muse Wonder |
| Terminal-Bench 2.1 (Vals AI) | 69.7% | Not measured | Only Muse Spark 1.2 measured |
| SWE-bench Multimodal (Vals AI) | Not measured | 45.6% | Only Muse Wonder measured |
| Vibe Code Bench 1-100 (Vals AI) | Not measured | 26.6% | Only Muse Wonder measured |
| SkillsBench (Vals AI) | 53.0% | Not measured | Only Muse Spark 1.2 measured |
| CyberBench (Vals AI) | 69.5% | 80.4% | Muse Wonder |
| SciCode | 57.4% | 60.3% | Muse Wonder |
| FrontierSWE | 12.0% | Not measured | Only Muse Spark 1.2 measured |
| WeirdML | 60.3% | Not measured | Only Muse Spark 1.2 measured |
| LMArena Text: coding | 1505 | 1522 | Muse Wonder |
| LMArena Text: software and IT | 1505 | 1515 | Muse Wonder |
| LMArena WebDev: React | 1526 | 1641 | Muse Wonder |
| LMArena WebDev: HTML | 1543 | 1694 | Muse Wonder |
| Neither model has a result on the other 22 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | Muse Spark 1.2 | Muse Wonder | Leader |
|---|---|---|---|
| AA Intelligence Index | 39.6 | Not measured | Only Muse Spark 1.2 measured |
| Epoch Capabilities Index | 155.1 | 161.8 | Muse Wonder |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index14
| Benchmark | Muse Spark 1.2 | Muse Wonder | Leader |
|---|---|---|---|
| AA Agentic Index | 43.2 | Not measured | Only Muse Spark 1.2 measured |
| Design Arena: agents building web apps | 1221 | Not measured | Only Muse Spark 1.2 measured |
| Design Arena: agents building full-stack apps | 1210 | Not measured | Only Muse Spark 1.2 measured |
| Design Arena: agents building mobile apps | 1181 | Not measured | Only Muse Spark 1.2 measured |
| Design Arena: agents building Android apps | 1222 | Not measured | Only Muse Spark 1.2 measured |
| Design Arena: agents building games | 1178 | Not measured | Only Muse Spark 1.2 measured |
| Design Arena: HTML slides | 1074 | Not measured | Only Muse Spark 1.2 measured |
| Design Arena: PowerPoint slides | 1170 | Not measured | Only Muse Spark 1.2 measured |
| Design Arena: spreadsheets | Not measured | 1258 | Only Muse Wonder measured |
| LiveBench research | 67.7% | Not measured | Only Muse Spark 1.2 measured |
| LiveBench browser | 36.0% | Not measured | Only Muse Spark 1.2 measured |
| LiveBench long agentic | 21.3% | Not measured | Only Muse Spark 1.2 measured |
| SREBench (Vals AI) | 0.0% | Not measured | Only Muse Spark 1.2 measured |
| Finance Agent (Vals AI) | 60.6% | 56.6% | Muse Spark 1.2 |
| Tax Agent Bench (Vals AI) | 32.8% | 36.3% | Muse Wonder |
| Legal Research Bench (Vals AI) | 43.8% | 50.5% | Muse Wonder |
| Harvey's Legal Agent Benchmark (Vals AI) | Not measured | 23.8% | Only Muse Wonder measured |
| Excel Modeling Benchmark (Vals AI) | 57.0% | 67.1% | Muse Wonder |
| BioMysteryBench (Vals AI) | 64.8% | 67.0% | Muse Wonder |
| Terminal-Bench-Science (Vals AI) | Not measured | 22.9% | Only Muse Wonder measured |
| Vals Science Index | Not measured | 39.1% | Only Muse Wonder measured |
| CUA-bench (Vals AI) | Not measured | 7.2% | Only Muse Wonder measured |
| APEX-Agents | 45.4% | Not measured | Only Muse Spark 1.2 measured |
| Vending-Bench 2 | Not measured | $9,432 | Only Muse Wonder measured |
| Neither model has a result on the other 18 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index04
| Benchmark | Muse Spark 1.2 | Muse Wonder | Leader |
|---|---|---|---|
| LiveBench reasoning | 90.0% | Not measured | Only Muse Spark 1.2 measured |
| MysteryMechanism (Vals AI) | Not measured | 27.5% | Only Muse Wonder measured |
| MMMU Pro (Vals AI) | 86.1% | Not measured | Only Muse Spark 1.2 measured |
| GPQA Diamond (Artificial Analysis run) | 90.4% | Not measured | Only Muse Spark 1.2 measured |
| Humanity's Last Exam (Artificial Analysis run) | 45.5% | 50.1% | Muse Wonder |
| SimpleBench | 74.5% | Not measured | Only Muse Spark 1.2 measured |
| Chess puzzles | 27.0% | Not measured | Only Muse Spark 1.2 measured |
| LMArena Text: hard prompts | 1489 | 1506 | Muse Wonder |
| LMArena Text: expert | 1475 | 1513 | Muse Wonder |
| LMArena Text: hard prompts in English | 1487 | 1506 | Muse Wonder |
| LMArena Vision: diagrams | 1317 | Not measured | Only Muse Spark 1.2 measured |
| Neither model has a result on the other 12 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index03
| Benchmark | Muse Spark 1.2 | Muse Wonder | Leader |
|---|---|---|---|
| ProofBench | 43.0% | 84.0% | Muse Wonder |
| LMArena Text: math | 1484 | 1486 | Muse Wonder |
| LMArena Text: mathematical work | 1480 | 1514 | Muse Wonder |
| Neither model has a result on the other 11 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index22
| Benchmark | Muse Spark 1.2 | Muse Wonder | Leader |
|---|---|---|---|
| MMLU Pro | 88.3% | Not measured | Only Muse Spark 1.2 measured |
| LegalBench (Vals AI) | 85.3% | Not measured | Only Muse Spark 1.2 measured |
| CorpFin (Vals AI) | 70.9% | Not measured | Only Muse Spark 1.2 measured |
| TaxEval (Vals AI) | 80.4% | Not measured | Only Muse Spark 1.2 measured |
| MortgageTax (Vals AI) | 65.4% | Not measured | Only Muse Spark 1.2 measured |
| MedCode (Vals AI) | 49.4% | Not measured | Only Muse Spark 1.2 measured |
| SimpleQA Verified | 60.3% | Not measured | Only Muse Spark 1.2 measured |
| LMArena Text: science | 1500 | 1506 | Muse Wonder |
| LMArena Text: medicine and healthcare | 1496 | 1512 | Muse Wonder |
| LMArena Text: legal and government | 1517 | 1505 | Muse Spark 1.2 |
| LMArena Text: business and finance | 1484 | 1483 | Muse Spark 1.2 |
| Neither model has a result on the other 9 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index15
| Benchmark | Muse Spark 1.2 | Muse Wonder | Leader |
|---|---|---|---|
| Design Arena: websites | 1312 | 1363 | Muse Wonder |
| Design Arena: UI components | 1299 | 1380 | Muse Wonder |
| Design Arena: game dev | 1300 | 1379 | Muse Wonder |
| Design Arena: data viz | 1334 | 1312 | Muse Spark 1.2 |
| Design Arena: 3D | 1300 | 1428 | Muse Wonder |
| Design Arena: all code categories | 1312 | 1372 | Muse Wonder |
| Design Arena: SVG | 1283 | Not measured | Only Muse Spark 1.2 measured |
| Design Arena: ASCII art | 1284 | Not measured | Only Muse Spark 1.2 measured |
| LiveBench design | 67.1% | Not measured | Only Muse Spark 1.2 measured |
GeneralLike-for-likePublic boards, shown as published, kept out of the index016
| Benchmark | Muse Spark 1.2 | Muse Wonder | Leader |
|---|---|---|---|
| Vals Index | Not measured | 64.9% | Only Muse Wonder measured |
| LiveBench data analysis | 76.5% | Not measured | Only Muse Spark 1.2 measured |
| Public Benefits Bench (Vals AI) | 68.5% | 70.2% | Muse Wonder |
| SAGE (Vals AI) | 47.7% | Not measured | Only Muse Spark 1.2 measured |
| MedScribe (Vals AI) | 90.1% | Not measured | Only Muse Spark 1.2 measured |
| Vals Multimodal Index | 69.8% | Not measured | Only Muse Spark 1.2 measured |
| Long context reasoning (AA-LCR) | 79.0% | 80.7% | Muse Wonder |
| DTBench | 94.7% | 95.2% | Muse Wonder |
| LMCA | 48.4% | 56.8% | Muse Wonder |
| LMArena Text | 1482 | 1502 | Muse Wonder |
| LMArena Text: instruction following | 1462 | 1491 | Muse Wonder |
| LMArena Text: creative writing | 1436 | 1491 | Muse Wonder |
| LMArena Vision | 1304 | Not measured | Only Muse Spark 1.2 measured |
| LMArena Text: English | 1483 | 1507 | Muse Wonder |
| LMArena Text: non-English | 1476 | 1492 | Muse Wonder |
| LMArena Text: Chinese | 1509 | 1537 | Muse Wonder |
| LMArena Text: Korean | 1462 | Not measured | Only Muse Spark 1.2 measured |
| LMArena Text: French | 1495 | 1523 | Muse Wonder |
| LMArena Text: Spanish | 1484 | Not measured | Only Muse Spark 1.2 measured |
| LMArena Text: Russian | 1483 | 1522 | Muse Wonder |
| LMArena Text: multi-turn | 1502 | 1510 | Muse Wonder |
| LMArena Text: longer queries | 1473 | 1493 | Muse Wonder |
| LMArena Text: writing and language | 1453 | 1496 | Muse Wonder |
| LMArena Text: entertainment and media | 1443 | 1482 | Muse Wonder |
| LMArena Vision: English | 1291 | Not measured | Only Muse Spark 1.2 measured |
| LMArena Vision: reading text | 1313 | Not measured | Only Muse Spark 1.2 measured |
| Neither model has a result on the other 22 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Muse Spark 1.2 | Muse Wonder | Leader |
|---|---|---|---|
| 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 costs less
Repository review
50K input + 3K output tokens
Muse Spark 1.2 costs less
Agent loop
200K input + 10K output tokens
Muse Spark 1.2 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.
| Specification | Meta Muse Spark 1.2 | Meta Muse Wonder |
|---|---|---|
| Maker | Meta | Meta |
| Weights | No public weights listed | No public weights listed |
| Released | August 5, 2026 | October 12, 2026 |
| Context window | 1.05M | 1.05M |
| Input $/Mtok | $1.25 | $4.00 |
| Output $/Mtok | $4.25 | $20.00 |
| Coding index | 55.7 | 71.9 |
| Independent measurements behind it | 5 | 3 |
| Cost per solved task | $0.64 | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Muse Spark 1.2 better than Muse Wonder for coding?
Pick Muse Wonder for coding: it scores 71.9 on the Vibe Coding Index against 55.7 for Muse Spark 1.2, though on 3 benchmarks against 5. Pick Muse Spark 1.2 for output tokens 4.7x cheaper ($4.25 vs $20.00 per million) and work measured by Finance Agent (Vals AI) or LMArena Text: legal and government, where it leads.
Which is cheaper, Muse Spark 1.2 or Muse Wonder?
Muse Spark 1.2 costs $1.25 per million input tokens and $4.25 per million output tokens; Muse Wonder costs $4.00 and $20.00. Cost per solved task is known for Muse Spark 1.2 only.
Which has the bigger context window, Muse Spark 1.2 or Muse Wonder?
Neither: both take 1.05M tokens.
How many benchmarks is the Muse Spark 1.2 vs Muse Wonder comparison based on?
50 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 Wonder's on 3. Where only one model has a result, the page says so instead of naming a winner.