Claude Haiku 5.5 vs Muse Spark 1.2
Claude Haiku 5.5 leads Muse Spark 1.2 on coding, 70.7 to 55.7 on the Vibe Coding Index, and Claude Haiku 5.5 costs 8.5x less per output token ($0.50 vs $4.25 per million). Updated October 22, 2026; 44 benchmarks have a result for both.
Claude Haiku 5.5 or Muse Spark 1.2: which to pick
Pick Claude Haiku 5.5 for coding: it scores 70.7 on the Vibe Coding Index against 55.7 for Muse Spark 1.2.
Pick Muse Spark 1.2 for a larger context window (1.05M against 1M tokens) and work measured by SciCode or Finance Agent (Vals AI), where it leads.
- Shared results
- 44
- Claude Haiku 5.5 only
- 19
- Muse Spark 1.2 only
- 53
- 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
Claude Haiku 5.5
Claude Haiku 5.5 scores 70.7 against 55.7.
5 benchmarks behind each score
Intelligence
AA Intelligence Index
Claude Haiku 5.5
Claude Haiku 5.5 leads, 43.4 against 39.6.
Same board, both measured
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 Spark 1.2
Muse Spark 1.2 leads, 1312 against 1262.
Same board, both measured
General
Vals Index
No call
Only Claude Haiku 5.5 has a Vals Index result, 58.7%.
Directional only
Long documents
Listed context window
Muse Spark 1.2
Muse Spark 1.2 takes 1.05M tokens against 1M.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Claude Haiku 5.5
Claude Haiku 5.5 $0.0003, Muse Spark 1.2 $0.0034 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
- Claude Haiku 5.5 70.7 from 5 benchmarks, Muse Spark 1.2 55.7 from 5 benchmarks.
- Widest category
- Math, 46.0 points apart on average over 1 shared benchmark.
- Widest single gap
- ProofBench: Claude Haiku 5.5 89.0% against Muse Spark 1.2 43.0%, 46.0 points apart on a 0 to 100 scale.
- Output price
- Claude Haiku 5.5 $0.50 against Muse Spark 1.2 $4.25 per million output tokens, 8.5x apart.
- Context window
- Claude Haiku 5.5 takes 1M 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.
| Category | Claude Haiku 5.5 | Bars | Muse Spark 1.2 | Ties |
|---|---|---|---|---|
| Coding | 11 | 1 | ||
| Intelligence | 2 | 0 | ||
| Agentic work | 7 | 3 | ||
| Reasoning | 1 | 2 | ||
| Math | 1 | 0 | ||
| Knowledge | 1 | 1 | ||
| Web design | 3 | 4 | ||
| General | 3 | 4 | ||
| All shared boards | 29 | 15 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index111
| Benchmark | Claude Haiku 5.5 | Muse Spark 1.2 | Leader |
|---|---|---|---|
| LiveBench coding | 91.4% | 78.2% | Claude Haiku 5.5 |
| LMArena WebDev | 1586 | 1534 | Claude Haiku 5.5 |
| AA Coding Index | Not measured | 72.2 | Only Muse Spark 1.2 measured |
| SWE-bench Verified (Vals AI) | Not measured | 86.6% | Only Muse Spark 1.2 measured |
| Terminal-Bench 4 (Vals AI) | 36.4% | 6.1% | Claude Haiku 5.5 |
| Vibe Code Bench (Vals AI) | 90.4% | 79.1% | Claude Haiku 5.5 |
| Code migration (Vals AI) | 44.6% | 29.9% | Claude Haiku 5.5 |
| ProgramBench (Vals AI) | 2.5% | 0.5% | Claude Haiku 5.5 |
| IOI (Vals AI) | 47.3% | 21.8% | Claude Haiku 5.5 |
| Terminal-Bench 2.1 (Vals AI) | Not measured | 69.7% | Only Muse Spark 1.2 measured |
| SWE-bench Multimodal (Vals AI) | 44.0% | Not measured | Only Claude Haiku 5.5 measured |
| Vibe Code Bench 1-100 (Vals AI) | 20.4% | Not measured | Only Claude Haiku 5.5 measured |
| SkillsBench (Vals AI) | Not measured | 53.0% | Only Muse Spark 1.2 measured |
| CyberBench (Vals AI) | 75.5% | 69.5% | Claude Haiku 5.5 |
| SciCode | 55.0% | 57.4% | Muse Spark 1.2 |
| FrontierCode | 46.4% | Not measured | Only Claude Haiku 5.5 measured |
| FrontierSWE | 43.8% | 12.0% | Claude Haiku 5.5 |
| WeirdML | Not measured | 60.3% | Only Muse Spark 1.2 measured |
| LMArena Text: coding | Not measured | 1505 | Only Muse Spark 1.2 measured |
| LMArena Text: software and IT | Not measured | 1505 | Only Muse Spark 1.2 measured |
| LMArena WebDev: React | 1605 | 1526 | Claude Haiku 5.5 |
| LMArena WebDev: HTML | 1569 | 1543 | Claude Haiku 5.5 |
| SciCode | 9.2% | Not measured | Only Claude Haiku 5.5 measured |
| Neither model has a result on the other 16 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | Claude Haiku 5.5 | Muse Spark 1.2 | Leader |
|---|---|---|---|
| AA Intelligence Index | 43.4 | 39.6 | Claude Haiku 5.5 |
| Epoch Capabilities Index | 159.3 | 155.1 | Claude Haiku 5.5 |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index73
| Benchmark | Claude Haiku 5.5 | Muse Spark 1.2 | Leader |
|---|---|---|---|
| AA Agentic Index | Not measured | 43.2 | Only Muse Spark 1.2 measured |
| Design Arena: agents building web apps | Not measured | 1221 | Only Muse Spark 1.2 measured |
| Design Arena: agents building full-stack apps | Not measured | 1210 | Only Muse Spark 1.2 measured |
| Design Arena: agents building mobile apps | Not measured | 1181 | Only Muse Spark 1.2 measured |
| Design Arena: agents building Android apps | Not measured | 1223 | Only Muse Spark 1.2 measured |
| Design Arena: agents building games | Not measured | 1178 | Only Muse Spark 1.2 measured |
| Design Arena: HTML slides | Not measured | 1085 | Only Muse Spark 1.2 measured |
| Design Arena: PowerPoint slides | Not measured | 1173 | Only Muse Spark 1.2 measured |
| LiveBench research | 81.9% | 67.7% | Claude Haiku 5.5 |
| LiveBench browser | 80.0% | 36.0% | Claude Haiku 5.5 |
| LiveBench long agentic | 57.6% | 21.3% | Claude Haiku 5.5 |
| SREBench (Vals AI) | 5.7% | 0.0% | Claude Haiku 5.5 |
| Finance Agent (Vals AI) | 54.1% | 60.6% | Muse Spark 1.2 |
| Tax Agent Bench (Vals AI) | 31.9% | 32.8% | Muse Spark 1.2 |
| Legal Research Bench (Vals AI) | 43.3% | 43.8% | Muse Spark 1.2 |
| Harvey's Legal Agent Benchmark (Vals AI) | 4.2% | Not measured | Only Claude Haiku 5.5 measured |
| Excel Modeling Benchmark (Vals AI) | 59.0% | 57.0% | Claude Haiku 5.5 |
| BioMysteryBench (Vals AI) | 70.4% | 64.8% | Claude Haiku 5.5 |
| Terminal-Bench-Science (Vals AI) | 10.0% | Not measured | Only Claude Haiku 5.5 measured |
| Vals Science Index | 38.5% | Not measured | Only Claude Haiku 5.5 measured |
| APEX-Agents | 56.1% | 45.4% | Claude Haiku 5.5 |
| Vending-Bench 2 | $3,602 | Not measured | Only Claude Haiku 5.5 measured |
| Neither model has a result on the other 19 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index12
| Benchmark | Claude Haiku 5.5 | Muse Spark 1.2 | Leader |
|---|---|---|---|
| LiveBench reasoning | 82.4% | 90.0% | Muse Spark 1.2 |
| MysteryMechanism (Vals AI) | 35.1% | Not measured | Only Claude Haiku 5.5 measured |
| MMMU Pro (Vals AI) | Not measured | 86.1% | Only Muse Spark 1.2 measured |
| GPQA Diamond (Artificial Analysis run) | Not measured | 90.4% | Only Muse Spark 1.2 measured |
| Humanity's Last Exam (Artificial Analysis run) | 44.4% | 45.5% | Muse Spark 1.2 |
| GPQA Diamond (Epoch AI run) | 89.6% | Not measured | Only Claude Haiku 5.5 measured |
| ARC-AGI | 89.0% | Not measured | Only Claude Haiku 5.5 measured |
| ARC-AGI-2 | 61.3% | Not measured | Only Claude Haiku 5.5 measured |
| SimpleBench | Not measured | 74.5% | Only Muse Spark 1.2 measured |
| Chess puzzles | 55.0% | 27.0% | Claude Haiku 5.5 |
| Mystery game puzzles | 30.0% | Not measured | Only Claude Haiku 5.5 measured |
| EBR-bench | 43.8% | Not measured | Only Claude Haiku 5.5 measured |
| LMArena Text: hard prompts | Not measured | 1489 | Only Muse Spark 1.2 measured |
| LMArena Text: expert | Not measured | 1475 | Only Muse Spark 1.2 measured |
| LMArena Text: hard prompts in English | Not measured | 1487 | Only Muse Spark 1.2 measured |
| LMArena Vision: diagrams | Not measured | 1317 | Only Muse Spark 1.2 measured |
| Neither model has a result on the other 6 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index10
| Benchmark | Claude Haiku 5.5 | Muse Spark 1.2 | Leader |
|---|---|---|---|
| OTIS Mock AIME | 98.9% | Not measured | Only Claude Haiku 5.5 measured |
| FrontierMath, tiers 1 to 3 | 75.1% | Not measured | Only Claude Haiku 5.5 measured |
| FrontierMath, tier 4 | 46.3% | Not measured | Only Claude Haiku 5.5 measured |
| ProofBench | 89.0% | 43.0% | Claude Haiku 5.5 |
| LMArena Text: math | Not measured | 1484 | Only Muse Spark 1.2 measured |
| LMArena Text: mathematical work | Not measured | 1480 | Only Muse Spark 1.2 measured |
| Neither model has a result on the other 7 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index11
| Benchmark | Claude Haiku 5.5 | Muse Spark 1.2 | Leader |
|---|---|---|---|
| MMLU Pro | Not measured | 88.3% | Only Muse Spark 1.2 measured |
| LegalBench (Vals AI) | Not measured | 85.3% | Only Muse Spark 1.2 measured |
| CorpFin (Vals AI) | Not measured | 70.9% | Only Muse Spark 1.2 measured |
| TaxEval (Vals AI) | Not measured | 80.4% | Only Muse Spark 1.2 measured |
| MortgageTax (Vals AI) | Not measured | 65.4% | Only Muse Spark 1.2 measured |
| MedCode (Vals AI) | 51.5% | 49.4% | Claude Haiku 5.5 |
| SimpleQA Verified | 23.8% | 60.3% | Muse Spark 1.2 |
| LMArena Text: science | Not measured | 1500 | Only Muse Spark 1.2 measured |
| LMArena Text: medicine and healthcare | Not measured | 1496 | Only Muse Spark 1.2 measured |
| LMArena Text: legal and government | Not measured | 1517 | Only Muse Spark 1.2 measured |
| LMArena Text: business and finance | Not measured | 1484 | Only Muse Spark 1.2 measured |
| Neither model has a result on the other 7 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index34
| Benchmark | Claude Haiku 5.5 | Muse Spark 1.2 | Leader |
|---|---|---|---|
| Design Arena: websites | 1262 | 1312 | Muse Spark 1.2 |
| Design Arena: UI components | 1283 | 1299 | Muse Spark 1.2 |
| Design Arena: game dev | 1315 | 1301 | Claude Haiku 5.5 |
| Design Arena: data viz | 1294 | 1334 | Muse Spark 1.2 |
| Design Arena: 3D | 1342 | 1300 | Claude Haiku 5.5 |
| Design Arena: all code categories | 1277 | 1312 | Muse Spark 1.2 |
| Design Arena: SVG | Not measured | 1283 | Only Muse Spark 1.2 measured |
| Design Arena: ASCII art | Not measured | 1285 | Only Muse Spark 1.2 measured |
| LiveBench design | 83.1% | 67.1% | Claude Haiku 5.5 |
GeneralLike-for-likePublic boards, shown as published, kept out of the index34
| Benchmark | Claude Haiku 5.5 | Muse Spark 1.2 | Leader |
|---|---|---|---|
| Vals Index | 58.7% | Not measured | Only Claude Haiku 5.5 measured |
| LiveBench data analysis | 53.1% | 76.5% | Muse Spark 1.2 |
| Public Benefits Bench (Vals AI) | 63.2% | 68.5% | Muse Spark 1.2 |
| SAGE (Vals AI) | 45.8% | 47.7% | Muse Spark 1.2 |
| MedScribe (Vals AI) | 89.5% | 90.1% | Muse Spark 1.2 |
| Vals Multimodal Index | Not measured | 69.8% | Only Muse Spark 1.2 measured |
| Long context reasoning (AA-LCR) | 82.7% | 79.0% | Claude Haiku 5.5 |
| DTBench | 95.0% | 94.7% | Claude Haiku 5.5 |
| LMCA | 51.1% | 48.4% | Claude Haiku 5.5 |
| Furniture assembly | 47.5% | Not measured | Only Claude Haiku 5.5 measured |
| LMArena Text | Not measured | 1482 | Only Muse Spark 1.2 measured |
| LMArena Text: instruction following | Not measured | 1462 | Only Muse Spark 1.2 measured |
| LMArena Text: creative writing | Not measured | 1436 | Only Muse Spark 1.2 measured |
| LMArena Vision | Not measured | 1304 | Only Muse Spark 1.2 measured |
| LMArena Text: English | Not measured | 1483 | Only Muse Spark 1.2 measured |
| LMArena Text: non-English | Not measured | 1476 | Only Muse Spark 1.2 measured |
| LMArena Text: Chinese | Not measured | 1509 | Only Muse Spark 1.2 measured |
| LMArena Text: Korean | Not measured | 1462 | Only Muse Spark 1.2 measured |
| LMArena Text: French | Not measured | 1495 | Only Muse Spark 1.2 measured |
| LMArena Text: Spanish | Not measured | 1484 | Only Muse Spark 1.2 measured |
| LMArena Text: Russian | Not measured | 1483 | Only Muse Spark 1.2 measured |
| LMArena Text: multi-turn | Not measured | 1502 | Only Muse Spark 1.2 measured |
| LMArena Text: longer queries | Not measured | 1473 | Only Muse Spark 1.2 measured |
| LMArena Text: writing and language | Not measured | 1453 | Only Muse Spark 1.2 measured |
| LMArena Text: entertainment and media | Not measured | 1443 | Only Muse Spark 1.2 measured |
| LMArena Vision: English | Not measured | 1291 | Only Muse Spark 1.2 measured |
| LMArena Vision: reading text | Not measured | 1313 | Only Muse Spark 1.2 measured |
| Neither model has a result on the other 19 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Claude Haiku 5.5 | Muse Spark 1.2 | 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
Claude Haiku 5.5 costs less
Repository review
50K input + 3K output tokens
Claude Haiku 5.5 costs less
Agent loop
200K input + 10K output tokens
Claude Haiku 5.5 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.
| Specification | Anthropic Claude Haiku 5.5 | Meta Muse Spark 1.2 |
|---|---|---|
| Maker | Anthropic | Meta |
| Weights | No public weights listed | No public weights listed |
| Released | October 7, 2026 | August 5, 2026 |
| Context window | 1M | 1.05M |
| Input $/Mtok | $0.10 | $1.25 |
| Output $/Mtok | $0.50 | $4.25 |
| Coding index | 70.7 | 55.7 |
| Independent measurements behind it | 5 | 5 |
| Cost per solved task | Not computable | $0.64 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Claude Haiku 5.5 better than Muse Spark 1.2 for coding?
Pick Claude Haiku 5.5 for coding: it scores 70.7 on the Vibe Coding Index against 55.7 for Muse Spark 1.2. Pick Muse Spark 1.2 for a larger context window (1.05M against 1M tokens) and work measured by SciCode or Finance Agent (Vals AI), where it leads.
Which is cheaper, Claude Haiku 5.5 or Muse Spark 1.2?
Claude Haiku 5.5 costs $0.10 per million input tokens and $0.50 per million output tokens; Muse Spark 1.2 costs $1.25 and $4.25. Cost per solved task is known for Muse Spark 1.2 only.
Which has the bigger context window, Claude Haiku 5.5 or Muse Spark 1.2?
Muse Spark 1.2, with 1.05M tokens against 1M.
How many benchmarks is the Claude Haiku 5.5 vs Muse Spark 1.2 comparison based on?
44 benchmarks have a result for both models as of October 22, 2026. Claude Haiku 5.5's coding index rests on 5 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.