Compare / Gemini 3.5 Flash vs Muse Spark 1.1
Gemini 3.5 Flash vs Muse Spark 1.1
20 results from 5 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | Gemini 3.5 Flash | Muse Spark 1.1 |
|---|---|---|---|
| Arena (formerly LMArena)reported | OverallArena rating | 1,461 | 1,479 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,476 | 1,463 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,468 | 1,493 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,466 | 1,449 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,491 | 1,504 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,465 | 1,473 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,477 | 1,491 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,472 | 1,460 |
| Artificial Analysisreported | Artificial Analysis Coding Indexindex score | 70.1 | 71.3 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 33.6 | 33.7 |
| Artificial Analysisreported | GPQA Diamond% of questions | 92.2% | 89.8% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 42.7% | 46.2% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 74.3% | 77.7% |
| Artificial Analysisreported | SciCode% of problems | 53.9% | 58.8% |
| Artificial Analysisreported | Terminal-Bench 2.1% of tasks | 78.7% | 77.9% |
| Artificial Analysisreported | Terminal-Bench 4.0% of tasks | 6.6% | 6.1% |
| Artificial Analysisreported | Τ-bench banking% of tasks | 32.2% | 31.8% |
| GDP.pdfreported | Professional document tasks% of rubric | 14.0% | 15.0% |
| SimpleQA Verified (Epoch AI)reported | Correct answers% of questions | 66.2% | 57.8% |
| Vending-Bench 2reported | Money after a yearUS dollars | $5,396.42 | $6,520.48 |
Bold green marks the better value on that metric. Where a source reports ranges that overlap, the difference may not be meaningful; see the benchmark page for ranges.