Muse Wonder vs Qwen2.5 7B Instruct
Qwen2.5 7B Instruct costs 100.0x less per output token than Muse Wonder ($0.20 vs $20.00 per million). Updated October 22, 2026; 3 benchmarks have a result for both.
Muse Wonder or Qwen2.5 7B Instruct: which to pick
Only Muse Wonder has a coding score, 71.9 on the Vibe Coding Index; Qwen2.5 7B Instruct has none, so the two cannot be ranked on coding.
Pick Qwen2.5 7B Instruct for output tokens 100.0x cheaper ($0.20 vs $20.00 per million).
- Shared results
- 3
- Muse Wonder only
- 57
- Qwen2.5 7B Instruct only
- 4
- Like-for-like categories
- 2 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
No call
Only Muse Wonder has a coding score, so no call is made.
One side not measured
Web design
Design Arena: websites
No call
Only Muse Wonder has a Design Arena: websites result, 1362.
Directional only
General
Vals Index
No call
Only Muse Wonder has a Vals Index result, 64.9%.
Directional only
Long documents
Listed context window
Muse Wonder
Muse Wonder takes 1.05M tokens against 33K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Qwen2.5 7B Instruct
Muse Wonder $0.014, Qwen2.5 7B Instruct $0.0002 per turn.
List prices
At a glance
- Coding index
- Muse Wonder 71.9 from 3 benchmarks, Qwen2.5 7B Instruct not benchmarked.
- Widest category
- Intelligence, 54.1 points apart on average over 1 shared benchmark.
- Widest single gap
- Epoch Capabilities Index: Muse Wonder 161.8 against Qwen2.5 7B Instruct 118.5, 54.1 points apart on a 0 to 100 scale.
- Output price
- Muse Wonder $20.00 against Qwen2.5 7B Instruct $0.20 per million output tokens, 100.0x apart.
- Context window
- Muse Wonder takes 1.05M tokens, Qwen2.5 7B Instruct 33K.
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 Wonder | Bars | Qwen2.5 7B Instruct | Ties |
|---|---|---|---|---|
| Intelligence | 1 | 0 | ||
| General | 2 | 0 | ||
| All shared boards | 3 | 0 |
CodingDirectional onlySix of these boards feed the Vibe Coding Index13 boards, none shared
| Benchmark | Muse Wonder | Qwen2.5 7B Instruct | Leader |
|---|---|---|---|
| LMArena WebDev | 1652 | Not measured | Only Muse Wonder measured |
| Terminal-Bench 4 (Vals AI) | 49.5% | Not measured | Only Muse Wonder measured |
| Vibe Code Bench (Vals AI) | 83.1% | Not measured | Only Muse Wonder measured |
| Code migration (Vals AI) | 62.5% | Not measured | Only Muse Wonder measured |
| IOI (Vals AI) | 66.9% | Not measured | Only Muse Wonder measured |
| SWE-bench Multimodal (Vals AI) | 45.6% | Not measured | Only Muse Wonder measured |
| Vibe Code Bench 1-100 (Vals AI) | 26.6% | Not measured | Only Muse Wonder measured |
| CyberBench (Vals AI) | 80.4% | Not measured | Only Muse Wonder measured |
| SciCode | 60.3% | Not measured | Only Muse Wonder measured |
| LMArena Text: coding | 1522 | Not measured | Only Muse Wonder measured |
| LMArena Text: software and IT | 1515 | Not measured | Only Muse Wonder measured |
| LMArena WebDev: React | 1641 | Not measured | Only Muse Wonder measured |
| LMArena WebDev: HTML | 1694 | Not measured | Only Muse Wonder measured |
| Neither model has a result on the other 26 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index10
| Benchmark | Muse Wonder | Qwen2.5 7B Instruct | Leader |
|---|---|---|---|
| Epoch Capabilities Index | 161.8 | 118.5 | Muse Wonder |
| Neither model has a result on the other 1 board here | |||
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index12 boards, none shared
| Benchmark | Muse Wonder | Qwen2.5 7B Instruct | Leader |
|---|---|---|---|
| Design Arena: spreadsheets | 1243 | Not measured | Only Muse Wonder measured |
| Finance Agent (Vals AI) | 56.6% | Not measured | Only Muse Wonder measured |
| Tax Agent Bench (Vals AI) | 36.3% | Not measured | Only Muse Wonder measured |
| Legal Research Bench (Vals AI) | 50.5% | Not measured | Only Muse Wonder measured |
| Harvey's Legal Agent Benchmark (Vals AI) | 23.8% | Not measured | Only Muse Wonder measured |
| Excel Modeling Benchmark (Vals AI) | 67.1% | Not measured | Only Muse Wonder measured |
| BioMysteryBench (Vals AI) | 67.0% | Not measured | Only Muse Wonder measured |
| Terminal-Bench-Science (Vals AI) | 22.9% | Not measured | Only Muse Wonder measured |
| Vals Science Index | 39.1% | Not measured | Only Muse Wonder measured |
| CUA-bench (Vals AI) | 7.2% | Not measured | Only Muse Wonder measured |
| BALROG | Not measured | 7.8% | Only Qwen2.5 7B Instruct measured |
| Vending-Bench 2 | $9,432 | Not measured | Only Muse Wonder measured |
| Neither model has a result on the other 29 boards here | |||
ReasoningDirectional onlyPublic boards, shown as published, kept out of the index6 boards, none shared
| Benchmark | Muse Wonder | Qwen2.5 7B Instruct | Leader |
|---|---|---|---|
| MysteryMechanism (Vals AI) | 27.5% | Not measured | Only Muse Wonder measured |
| Humanity's Last Exam (Artificial Analysis run) | 50.1% | Not measured | Only Muse Wonder measured |
| GPQA Diamond (Epoch AI run) | Not measured | 35.5% | Only Qwen2.5 7B Instruct measured |
| LMArena Text: hard prompts | 1506 | Not measured | Only Muse Wonder measured |
| LMArena Text: expert | 1513 | Not measured | Only Muse Wonder measured |
| LMArena Text: hard prompts in English | 1506 | Not measured | Only Muse Wonder measured |
| Neither model has a result on the other 16 boards here | |||
MathDirectional onlyPublic boards, shown as published, kept out of the index4 boards, none shared
| Benchmark | Muse Wonder | Qwen2.5 7B Instruct | Leader |
|---|---|---|---|
| OTIS Mock AIME | Not measured | 2.5% | Only Qwen2.5 7B Instruct measured |
| ProofBench | 84.0% | Not measured | Only Muse Wonder measured |
| LMArena Text: math | 1486 | Not measured | Only Muse Wonder measured |
| LMArena Text: mathematical work | 1514 | Not measured | Only Muse Wonder measured |
| Neither model has a result on the other 9 boards here | |||
KnowledgeDirectional onlyPublic boards, shown as published, kept out of the index5 boards, none shared
| Benchmark | Muse Wonder | Qwen2.5 7B Instruct | Leader |
|---|---|---|---|
| MMLU | Not measured | 72.9% | Only Qwen2.5 7B Instruct measured |
| LMArena Text: science | 1506 | Not measured | Only Muse Wonder measured |
| LMArena Text: medicine and healthcare | 1512 | Not measured | Only Muse Wonder measured |
| LMArena Text: legal and government | 1505 | Not measured | Only Muse Wonder measured |
| LMArena Text: business and finance | 1483 | Not measured | Only Muse Wonder measured |
| Neither model has a result on the other 13 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index6 boards, none shared
| Benchmark | Muse Wonder | Qwen2.5 7B Instruct | Leader |
|---|---|---|---|
| Design Arena: websites | 1362 | Not measured | Only Muse Wonder measured |
| Design Arena: UI components | 1379 | Not measured | Only Muse Wonder measured |
| Design Arena: game dev | 1376 | Not measured | Only Muse Wonder measured |
| Design Arena: data viz | 1317 | Not measured | Only Muse Wonder measured |
| Design Arena: 3D | 1427 | Not measured | Only Muse Wonder measured |
| Design Arena: all code categories | 1371 | Not measured | Only Muse Wonder measured |
| Neither model has a result on the other 3 boards here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | Muse Wonder | Qwen2.5 7B Instruct | Leader |
|---|---|---|---|
| Vals Index | 64.9% | Not measured | Only Muse Wonder measured |
| Public Benefits Bench (Vals AI) | 70.2% | Not measured | Only Muse Wonder measured |
| Long context reasoning (AA-LCR) | 80.7% | Not measured | Only Muse Wonder measured |
| DTBench | 95.2% | 47.7% | Muse Wonder |
| LMCA | 56.8% | 6.4% | Muse Wonder |
| LMArena Text | 1502 | Not measured | Only Muse Wonder measured |
| LMArena Text: instruction following | 1491 | Not measured | Only Muse Wonder measured |
| LMArena Text: creative writing | 1491 | Not measured | Only Muse Wonder measured |
| LMArena Text: English | 1507 | Not measured | Only Muse Wonder measured |
| LMArena Text: non-English | 1492 | Not measured | Only Muse Wonder measured |
| LMArena Text: Chinese | 1537 | Not measured | Only Muse Wonder measured |
| LMArena Text: French | 1523 | Not measured | Only Muse Wonder measured |
| LMArena Text: Russian | 1522 | Not measured | Only Muse Wonder measured |
| LMArena Text: multi-turn | 1510 | Not measured | Only Muse Wonder measured |
| LMArena Text: longer queries | 1493 | Not measured | Only Muse Wonder measured |
| LMArena Text: writing and language | 1496 | Not measured | Only Muse Wonder measured |
| LMArena Text: entertainment and media | 1482 | Not measured | Only Muse Wonder measured |
| Neither model has a result on the other 29 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Muse Wonder | Qwen2.5 7B Instruct | 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
Qwen2.5 7B Instruct costs less
Repository review
50K input + 3K output tokens
Qwen2.5 7B Instruct costs less
Agent loop
200K input + 10K output tokens
Qwen2.5 7B Instruct 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 Wonder | Qwen2.5 7B Instruct |
|---|---|---|
| Maker | Meta | Qwen |
| Weights | No public weights listed | Open weights |
| Released | October 12, 2026 | October 16, 2024 |
| Context window | 1.05M | 33K |
| Input $/Mtok | $4.00 | $0.10 |
| Output $/Mtok | $20.00 | $0.20 |
| Coding index | 71.9 | Not benchmarked |
| Independent measurements behind it | 3 | 0 |
| Cost per solved task | Not computable | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Muse Wonder better than Qwen2.5 7B Instruct for coding?
Only Muse Wonder has a coding score, 71.9 on the Vibe Coding Index; Qwen2.5 7B Instruct has none, so the two cannot be ranked on coding. Pick Qwen2.5 7B Instruct for output tokens 100.0x cheaper ($0.20 vs $20.00 per million).
Which is cheaper, Muse Wonder or Qwen2.5 7B Instruct?
Muse Wonder costs $4.00 per million input tokens and $20.00 per million output tokens; Qwen2.5 7B Instruct costs $0.10 and $0.20. Cost per solved task is known for neither of them.
Which has the bigger context window, Muse Wonder or Qwen2.5 7B Instruct?
Muse Wonder, with 1.05M tokens against 33K.
How many benchmarks is the Muse Wonder vs Qwen2.5 7B Instruct comparison based on?
3 benchmarks have a result for both models as of October 22, 2026. Muse Wonder's coding index rests on 3 independent benchmarks and Qwen2.5 7B Instruct's on 0. Where only one model has a result, the page says so instead of naming a winner.