Comparing 2 models
Gemini 3.1 Pro Preview vs Muse Spark 1.3
Benchmarks through
Verdict
Gemini 3.1 Pro Preview places higher on 9 of the 9 tasks where both are ranked. Muse Spark 1.3 costs 1.6x less per input token.
- Choose Gemini 3.1 Pro Preview for Chat & Roleplay (#3), Roleplay & Character (#3), Creative Writing (#3).
Built from the published task rankings below; ranks come from each task's own weighted benchmarks, so a win on one task says nothing about another.
Measured head to head
Benchmarks both models were run on
| Evaluation | Gemini 3.1 Pro Preview | Muse Spark 1.3 | Source |
|---|---|---|---|
| Design Arena Coding | 1220 Elo | 1351 Elo | source · 2026-11-15 |
| Design Arena Full-stack Agent | 1041 Elo | 1269 Elo | source · 2026-11-15 |
| Design Arena UI | 1215 Elo | 1346 Elo | source · 2026-11-15 |
| Design Arena Web-app Agent | 1103 Elo | 1273 Elo | source · 2026-11-15 |
| Design Arena Website | 1225 Elo | 1352 Elo | source · 2026-11-15 |
| FrontierMath Tiers 1-3 (Epoch AI) | 59.6% ± 2.9 | 74.4% ± 2.6 | source · 2026-06-11 |
| LiveBench Agentic Coding | 44.1% | 64.1% | source · 2026-10-18 |
| LiveBench Coding | 76.5% | 81.1% | source · 2026-10-18 |
| LiveBench Data Analysis | 78.5% | 79.6% | source · 2026-10-18 |
| LiveBench Instruction Following | 79.1% | 78% | source · 2026-10-18 |
| LiveBench Language | 85.4% | 82.8% | source · 2026-10-18 |
| LiveBench Math | 91.0% | 96.0% | source · 2026-10-18 |
| LiveBench Reasoning | 84% | 89.7% | source · 2026-10-18 |
| LMArena Agent | -4.5 | 2.2 | source · 2026-11-11 |
| LMArena Agent Bash Recovery | -23.3 | 6.1 | source · 2026-10-08 |
| LMArena Agent Steerability | -8.2 | 2.6 | source · 2026-11-11 |
| LMArena Agent Task Outcome | -11.3 | 3.7 | source · 2026-11-11 |
| LMArena Agent Tool Hallucination | -0.4 | 0.8 | source · 2026-11-11 |
| LMArena Creative Writing | 1483 Elo | 1467 Elo | source · 2026-11-11 |
| LMArena Instruction Following | 1465 Elo | 1479 Elo | source · 2026-11-11 |
| LMArena Long Query | 1482 Elo | 1489 Elo | source · 2026-11-11 |
| LMArena Math | 1485 Elo | 1498 Elo | source · 2026-11-11 |
| LMArena Text | 1487 Elo | 1497 Elo | source · 2026-11-11 |
| LMArena Text Factuality | 1472 Elo | 1481 Elo | source · 2026-11-11 |
| LMArena Vision | 1280 Elo | 1291 Elo | source · 2026-11-09 |
| LMArena Vision Diagrams | 1309 Elo | 1329 Elo | source · 2026-11-09 |
| LMArena Vision OCR | 1295 Elo | 1309 Elo | source · 2026-11-09 |
| LMArena WebDev | 1446 Elo | 1655 Elo | source · 2026-11-11 |
| LMArena WebDev Image-to-Web | 1480 Elo | 1637 Elo | source · 2026-11-10 |
| OTIS Mock AIME (Epoch AI) | 95.6% ± 3.1 | 99.2% ± 0.6 | source · 2026-08-06 |
| OTIS Mock AIME (Epoch AI) | 95.6% ± 3.1 | 99.2% ± 0.6 | source · 2026-02-20 |
Values are as published by each evaluator, in its own units; rows from different evaluators are not on a common scale.
task evidence
How they place when the question changes
Lower ranks are better. “Best” marks the strongest option in this shortlist for that task.