Muse Spark Open vs Muse Wonder Full
As of 18 Oct 2026, Muse Wonder Full (max) is the better pick for most workloads: it leads on quality (68.6 vs 61.2 on the BenchLeader Index).
Muse Spark Open Key differences
- Muse Wonder Full (max) leads on quality: 68.6 vs 61.2.
- Muse Spark Open (xhigh) is stronger in knowledge, multimodal.
- Muse Wonder Full (max) is stronger in composite, long context, reasoning, coding.
Choose Muse Spark Open (xhigh) if
- knowledge is the priority (73 vs 71 on the category score)
- you want to run it yourself: its weights are open
Choose Muse Wonder Full (max) if
- reasoning is the priority (95 vs 66 on the category score)
- long context is the priority (67 vs 64 on the category score)
Category by category
Category scores on the BenchLeader scale: 50 is the average evaluated model, 15 points is one standard deviation. A gap under one point is called level.
| Category | Muse Spark Open (xhigh) | Muse Wonder Full (max) | Leader |
|---|---|---|---|
| Reasoning | 65.5 | 95.0 | Muse Wonder Full |
| Coding | – | 64.0 | not comparable |
| Knowledge | 72.5 | 71.3 | Muse Spark Open |
| Multimodal | 64.1 | – | not comparable |
| Long context | 64.2 | 67.0 | Muse Wonder Full |
| Composite | 73.1 | 92.2 | Muse Wonder Full |
Price, speed and latency
List prices blended at three input tokens per output token. Per-task prices use the token counts on the methodology page; suite cost is what Artificial Analysis paid to run its whole Intelligence Index.
| Measure | Muse Spark Open (xhigh) | Muse Wonder Full (max) |
|---|---|---|
| Blended price per 1M tokens | $2.00 | – |
| Input / output per 1M | $1.25 / $4.25 | – / – |
| Price per task: chat reply | $0.0018 | – |
| Price per task: summarise a 30-page report | $0.018 | – |
| Price per task: code edit | $0.014 | – |
| Price per task: agentic coding session | $0.092 | – |
| Price per task: structured extraction | $0.0034 | – |
| Cost to run a full benchmark suite | $0.844 | – |
| Output speed | 140 tok/s | – |
| Time to first answer | 15.0 s | – |
| Time to a complete answer | 18.6 s | – |
Every benchmark
| Metric | Muse Spark Open (xhigh) | Muse Wonder Full (max) |
|---|---|---|
| BenchLeader Index | 61.2 | 68.6 |
| Composite score | 73.1 | 92.2 |
| Knowledge score | 72.5 | 71.3 |
| Long context score | 64.2 | 67.0 |
| Multimodal score | 64.1 | – |
| Reasoning score | 65.5 | 95.0 |
| Coding score | – | 64.0 |
| Blended price $/M | $2.00 | – |
| Output speed | 140 tok/s | – |
| Time to first answer | 15.0 s | – |
| Context window | 1.0M | 1M |
| AA Intelligence Index v4.3.2 | 38.9 | 55.8 |
| AA-LCR | 78.3% | 83.7% |
| MMMU-Pro | 80.9% | – |
| AA-Omniscience | 24.6 | 30.1 |
| GPQA Diamond (AA) | 89.0% | – |
| Humanity's Last Exam (AA) | 43.9% | 53.6% |
| SciCode (AA) | 58.5% | 62.6% |
| IOI | – | 79.8% |
| LegalBench | – | 87.4% |
| Vals Index | 46.6 | 64.6 |
| CritPt | 14.9% | 32.9% |
| GDPval-AA v2.1 | 45.1% | 67.8% |
| BioMysteryBench | – | 73.3% |
| Code Migration | 21.2% | 61.0% |
| CyberBench | – | 73.6% |
| Excel Modeling Benchmark | 53.7% | 71.9% |
| Finance Agent v2 | 61.3% | 59.1% |
| Harvey's Legal Agent Benchmark | 16.7% | 21.7% |
| Legal Research Bench | 38.5% | 53.4% |
| MedCode | – | 49.9% |
| MedScribe | – | 89.3% |
| MysteryMechanism | – | 35.1% |
| ProgramBench | – | 0.5% |
| Public Benefits Bench | – | 73.8% |
| SREBench | – | 11.8% |
| Tax Agent Bench | 26.3% | 43.0% |
| Terminal-Bench 4.0 (Vals) | 6.6% | 58.1% |
| Terminal-Bench Science | – | 37.1% |
| Vibe Code Bench 1-100 | – | 34.0% |
| Vibe Code Bench v1.1 | 76.2% | 86.0% |
| Terminal-Bench 4.0 (AA) | 8.1% | 57.1% |
| Terminal-Bench 2.1 (AA) | 76.4% | – |
| AutomationBench | 49.7% | 71.3% |
| GDP.pdf | 18.6% | 32.6% |
| Harvey LAB-AA v1.1 | – | 13.1% |
| AA-Omniscience: accuracy | 39.3% | 43.2% |
| AA-Omniscience: non-hallucination | 75.9% | 77.0% |
| AA-Briefcase v1.1 | 1279 | 1729 |
| AA Openness Index | 27.8 | – |
Data as of 2026-10-18. Best configuration of each model; every score links to its source on the model pages.
Muse Spark Open vs Muse Wonder Full: questions
- Is Muse Spark Open better than Muse Wonder Full?
- Muse Wonder Full (max) leads on quality: 68.6 vs 61.2. The BenchLeader Index combines every independent quality benchmark; Muse Wonder Full (max) is ahead overall as of 2026-10-18, but check the category scores for your use.
- Which has the larger context window?
- Muse Spark Open accepts more context: 1.0M against 1M tokens.