Gemini 3.1 Pro vs Muse Spark
Full profiles: Gemini 3.1 Pro (Google) and Muse Spark (Meta): every benchmark, price and rank.
Updated · Data from 11 sources
As of 6 Nov 2026, Gemini 3.1 Pro and Muse Spark are neck and neck on quality (60.5 vs 64.1 on the BenchLeader Index, inside the ±5.3 margin). Choose Gemini 3.1 Pro for long context (66 vs 64) and maths (61 vs 59); choose Muse Spark for human preference (68 vs 58), instruction following (77 vs 67), and coding (65 vs 57).
Key facts (as of 6 Nov 2026)
- Gemini 3.1 Pro scores 60.5 on the BenchLeader Index and Muse Spark scores 64.1 (BenchLeader, 6 Nov 2026).
- Gemini 3.1 Pro costs $4.50 per 1M tokens blended (list price, 6 Nov 2026).
- Of the 33 benchmarks both models have, Gemini 3.1 Pro leads 24 (BenchLeader, 6 Nov 2026).
Gemini 3.1 Pro benchmarks 
Muse Spark benchmarks Which to pick for…
The pick's numbers come first. Category scores on the BenchLeader scale (50 is the average model, 15 points one standard deviation). High confidence means a lead of half a standard deviation or more; under one point is called either.
| Use case | Pick | Why | Confidence |
|---|---|---|---|
| Coding | Muse Spark | 65 vs 57 category score; Arena Coding 1529 vs 1521 | High |
| Agents & tool use | Muse Spark | 65 vs 57 category score; Terminal-Bench Hard 45.5% vs 53.8% | High |
| Maths | Gemini 3.1 Pro | 61 vs 59 category score; Arena Maths 1488 vs 1466 | Medium |
| Reasoning | Muse Spark | 68 vs 66 category score; GPQA Diamond 89.8% vs 94.1% | Medium |
| Knowledge | Muse Spark | 63 vs 56 category score; PRBench Finance 52.4% vs 41.9% | Medium |
| Long documents | Gemini 3.1 Pro | 66 vs 64 category score; AA-LCR v1.1 82.0% vs 78.0% | Medium |
| Images and other media | Either | 65 vs 64 category score; Arena Vision 1296 vs 1306 | Either |
| Following instructions | Muse Spark | 77 vs 67 category score; MultiChallenge 75.5% vs 71.4% | High |
| Chat quality (human preference) | Muse Spark | 68 vs 58 on Arena votes | High |
| High volume on a budget | Neither flagship | A smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs | – |
Key differences
- Gemini 3.1 Pro and Muse Spark are neck and neck on quality (60.5 vs 64.1, within the ±5.3 margin).
- Gemini 3.1 Pro is stronger in long context, maths, multimodal.
- Muse Spark is stronger in agents & tools, coding, human preference, instruction following, knowledge, reasoning.
At a glance
| Spec | Gemini 3.1 Pro | Muse Spark |
|---|---|---|
| Developer | Meta | |
| Released | 19 Feb 2026 | 8 Apr 2026 |
| Context window (tokens) | 1,048,576 | 262,144 |
| Compared at | Gemini 3.1 Pro | Muse Spark |
| Open weights | No | No |
| Price in / out / cached input per 1M | $2.00 / $12.00 / $0.200 | – |
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 | Gemini 3.1 Pro | Muse Spark | Leader |
|---|---|---|---|
| Reasoning | 66.3 | 67.5 | Muse Spark |
| Coding | 56.7 | 64.9 | Muse Spark |
| Agents & tools | 57.1 | 64.9 | Muse Spark |
| Maths | 60.7 | 58.9 | Gemini 3.1 Pro |
| Knowledge | 55.8 | 62.9 | Muse Spark |
| Instruction following | 66.9 | 76.6 | Muse Spark |
| Human preference | 58.4 | 68.1 | Muse Spark |
| Multimodal | 64.5 | 64.2 | level |
| Long context | 65.9 | 63.8 | Gemini 3.1 Pro |
On one chart
Each domain on the index's own scale, where 50 is an average model. Click a domain to see the benchmarks behind it. Pick up to 5 models from the list.
ModelIndex
Where sources disagree
Across the 33 benchmarks both have, Gemini 3.1 Pro leads 24, Muse Spark leads 9. Every overall index that rates both puts Muse Spark ahead.
| Overall index | Gemini 3.1 Pro | Muse Spark | Leader |
|---|---|---|---|
| BenchLeader Index | 60.5 | 64.1 | neck and neck |
| Arena Text | 1487 | 1489 | Muse Spark |
Each publisher runs its own tasks, harnesses and effort settings, so their overall verdicts can differ; BenchLeader reconciles them on one scale. † measured at another effort setting of the same model (the publisher did not test the one compared here).
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. Cost, tokens and time per benchmark task are measured: Artificial Analysis’ averages over its whole Intelligence Index run.
| Measure | Gemini 3.1 Pro | Muse Spark |
|---|---|---|
| Blended price per 1M tokens | $4.50 | – |
| Input / output per 1M | $2.00 / $12.00 | – / – |
| Cached input per 1M | $0.200 | – |
| Price per task: chat reply | $0.0044 | – |
| Price per task: summarize a 30-page report | $0.031 | – |
| Price per task: code edit | $0.030 | – |
| Price per task: agentic coding session | $0.168 | – |
| Price per task: structured extraction | $0.0064 | – |
| Agentic coding session, input mostly cached | $0.087 | – |
| Output speed | 122 tok/s | – |
| Time to first answer | 18.2 s | – |
| Time to a complete answer | 43.8 s | – |
Each model is compared at its best-ranked reasoning-effort setting. Gemini 3.1 Pro scores 60.5 and takes 18 s to its first answer token; Gemini 3.1 Pro (high) scores 60.1 in 18 s. Muse Spark scores 64.1. Higher effort means more thinking before the answer: better scores, longer waits and more tokens billed. How reasoning effort changes scores and cost.
Every benchmark
147 benchmarks, grouped by category; each name links to the board and its leaders. † measured at another effort setting of the same model, shown for reference and not counted as a win.
| Metric | Publisher | Gemini 3.1 Pro | Muse Spark |
|---|---|---|---|
| BenchLeader Index | BenchLeader | 60.5 | 64.1 |
| Agents & tools score | BenchLeader | 57.1 | 64.9 |
| Coding score | BenchLeader | 56.7 | 64.9 |
| Human preference score | BenchLeader | 58.4 | 68.1 |
| Instruction following score | BenchLeader | 66.9 | 76.6 |
| Knowledge score | BenchLeader | 55.8 | 62.9 |
| Long context score | BenchLeader | 65.9 | 63.8 |
| Maths score | BenchLeader | 60.7 | 58.9 |
| Multimodal score | BenchLeader | 64.5 | 64.2 |
| Reasoning score | BenchLeader | 66.3 | 67.5 |
| Blended price $/M | BenchLeader | $4.50 | – |
| Output speed | BenchLeader | 122 tok/s | – |
| Time to first answer | BenchLeader | 18.2 s | – |
| Context window | BenchLeader | 1,048,576 tokens | 262,144 tokens |
| Reasoning | |||
| GPQA Diamond | Epoch AI Benchmarking Hub | 94.1% | 89.8% |
| Humanity's Last Exam | Scale AI / CAIS | 46.4% | 40.6% |
| SimpleBench | SimpleBench | 79.6% | – |
| Arena Hard Prompts | Arena (LMArena) | 1508 | 1506 |
| LiveBench 1.0 Reasoning | LiveBench | 84.0%† | – |
| GPQA Diamond (AA) | Artificial Analysis | 94.1% | 88.4% |
| Humanity's Last Exam (AA) | Artificial Analysis | 47.0% | 40.7% |
| HLE-Diamond (AA) | Artificial Analysis | 31.4% | – |
| GPQA Diamond (Vals) | Vals AI | 95.5%† | 89.7% |
| EnigmaEval | Scale AI SEAL | 36.8%† | – |
| ARC-AGI-1 | ARC Prize | 98.0% | – |
| ARC-AGI-2 | ARC Prize | 77.1% | – |
| ARC-AGI-3 | ARC Prize | 0.4% | – |
| CritPt | Artificial Analysis | 17.7% | 11.3% |
| MysteryMechanism | Vals AI | 20.7%† | – |
| Chess Puzzles | Epoch AI Benchmarking Hub | 51.0%† | – |
| Mystery Game Puzzles | Epoch AI Benchmarking Hub | 34.0%† | – |
| CL-bench | Epoch AI Benchmarking Hub | 20.8% | – |
| CL-bench Life | Epoch AI Benchmarking Hub | 16.9% | – |
| LMCA | Epoch AI Benchmarking Hub | 53.8%† | – |
| DTBench | Epoch AI Benchmarking Hub | 95.5%† | – |
| ForecastBench | ForecastBench | 64.3% | – |
| Coding | |||
| SciCode | SciCode | 58.7% | 51.5% |
| WeirdML | WeirdML | 72.1% | – |
| GSO-Bench | GSO-Bench | 22.6% | – |
Show all 147 benchmarks
| Metric | Publisher | Gemini 3.1 Pro | Muse Spark |
|---|---|---|---|
| Coding | |||
| Arena Coding | Arena (LMArena) | 1521 | 1529 |
| Arena WebDev | Arena (LMArena) | 1447 | – |
| LiveBench 1.0 Coding | LiveBench | 76.5%† | – |
| SciCode (AA) | Artificial Analysis | 58.7% | – |
| LiveCodeBench | Vals AI | 88.5%† | – |
| IOI | Vals AI | 51.8%† | – |
| SWE-bench (Vals) | Vals AI | 78.8%† | 74.4% |
| SWE-Bench Pro | Scale AI SEAL | 46.1%† | 55.0% |
| Code Migration | Vals AI | 17.3%† | – |
| ProgramBench | Vals AI | 0.0%† | – |
| Vibe Code Bench 1-100 | Vals AI | 6.7%† | – |
| Vibe Code Bench v1.1 | Vals AI | 32.0%† | 19.7% |
| SWE-Bench Pro (private) | Scale AI SEAL | 46.1%† | 55.0% |
| SWE Atlas: Codebase QnA | Scale AI SEAL | 13.5% | 24.2% |
| SWE Atlas: Refactoring | Scale AI SEAL | 33.8% | – |
| SWE Atlas: Test Writing | Scale AI SEAL | 29.8% | 31.1% |
| MirrorCode | Epoch AI Benchmarking Hub | 8.9%† | – |
| DeepSWE v1.1 | Epoch AI Benchmarking Hub | 11.7% | – |
| ALE-Bench | Epoch AI Benchmarking Hub | 1160.6 | – |
| AlgoTune | AlgoTune | 2.0 | – |
| Terminal-Bench 4.0 (AA) | Artificial Analysis | 4.0% | – |
| Terminal-Bench 2.1 (AA) | Artificial Analysis | 73.8% | 62.2% |
| Agents & tools | |||
| Terminal-Bench | Terminal-Bench | 80.2% | – |
| APEX-Agents | Mercor | 35.1% | – |
| Arena Agent | Arena (LMArena) | -4.4 | – |
| LiveBench Agentic Coding | LiveBench | 44.1%† | – |
| Terminal-Bench Hard | Artificial Analysis | 53.8% | 45.5% |
| τ²-Bench Telecom (AA) | Artificial Analysis | 95.6% | 91.5% |
| Terminal-Bench 2.1 (Vals) | Vals AI | 70.8%† | – |
| MCP Atlas | Scale AI SEAL | 78.2%† | 82.2% |
| HiL-Bench | Scale AI SEAL | 35.3% | – |
| τ³-bench Banking | τ²-bench | 26.0%† | – |
| GDPval-AA v2.1, normalised score | Artificial Analysis | 14.6% | – |
| GDPval-AA v2.1 | Artificial Analysis | 799 | – |
| τ³-Banking (AA) | Artificial Analysis | 21.4% | – |
| ITBench SRE (AA) | Artificial Analysis | 30.3% | – |
| Analyst Agent (AA) | Artificial Analysis | 41.3% | – |
| APEX-Agents (AA) | Artificial Analysis | 32.0% | – |
| Finance Agent v2 | Vals AI | 43.0%† | – |
| Harvey's Legal Agent Benchmark v1 | Vals AI | 0.0%† | – |
| Legal Research Bench | Vals AI | 20.7%† | – |
| SREBench | Vals AI | 0.0%† | – |
| Tax Agent Bench | Vals AI | 9.3%† | – |
| Terminal-Bench 2.0 (Vals) | Vals AI | 67.4%† | 59.5% |
| Terminal-Bench 4.0 (Vals) | Vals AI | 2.5%† | – |
| Terminal-Bench Science | Vals AI | 1.4%† | – |
| BALROG | BALROG | 57.0% | – |
| DeepResearch Bench | DeepResearch Bench | 47.8%† | – |
| PostTrainBench | Epoch AI Benchmarking Hub | 22.0% | – |
| ExploitBench | Epoch AI Benchmarking Hub | 26.1% | – |
| METR Time Horizons | METR | 77.0% | – |
| GBAEval | Epoch AI Benchmarking Hub | 0.8% | – |
| Vending-Bench 2 | Andon Labs | 911.2 | – |
| AutomationBench | Artificial Analysis | 35.4% | – |
| GDP.pdf (AA) | Artificial Analysis | 17.8% | – |
| Harvey LAB-AA v1.1 | Harvey | 0.0% | – |
| EnterpriseOps-Gym | Artificial Analysis | 42.2% | – |
| AA-Briefcase v1.1 | Artificial Analysis | 457 | – |
| BrowseComp | OpenAI | 31.2% | – |
| DeepSearchQA | 60.2% | – | |
| FACTS Search | 90.2% | – | |
| Maths | |||
| FrontierMath Tiers 1–3 | Epoch AI Benchmarking Hub | 59.6% | – |
| FrontierMath Tier 4 | Epoch AI Benchmarking Hub | 26.8% | – |
| OTIS Mock AIME | Epoch AI Benchmarking Hub | 95.6% | 88.9% |
| ProofBench | Vals AI | 26.0% | 17.0% |
| LiveBench Mathematics | LiveBench | 91.0%† | – |
| AIME (Vals) | Vals AI | 98.1%† | 96.9% |
| AIME 2026 | MathArena | 98.3% | – |
| HMMT February 2026 | MathArena | 94.7% | – |
| USAMO 2026 | MathArena | 74.4% | – |
| MathArena Apex | MathArena | 60.9% | – |
| Arena Maths | Arena (LMArena) | 1488 | 1466 |
| Knowledge | |||
| SimpleQA Verified | Epoch AI Benchmarking Hub | 73.5%† | – |
| LiveBench 1.0 Data Analysis | LiveBench | 78.5%† | – |
| MMLU-Pro | Vals AI | 91.0%† | 87.3% |
| LegalBench | Vals AI | 87.4%† | 84.2% |
| CorpFin | Vals AI | 64.5%† | 65.1% |
| TaxEval | Vals AI | 72.9%† | 77.7% |
| MedQA | Vals AI | 96.4%† | – |
| PRBench Finance | Scale AI SEAL | 41.9% | 52.4% |
| PRBench Legal | Scale AI SEAL | 44.0% | 52.3% |
| MultiNRC | Scale AI SEAL | 64.7% | 59.0% |
| BioMysteryBench | Vals AI | 60.0%† | – |
| CaseLaw v2 | Vals AI | 64.8%† | 63.1% |
| Excel Modeling Benchmark | Vals AI | 52.6%† | – |
| MedCode | Vals AI | 59.1%† | 51.3% |
| MedScribe | Vals AI | 76.1%† | 85.9% |
| Public Benefits Bench v1.1 | Vals AI | 53.8%† | – |
| DrugDiscoveryBench | Scale AI SEAL | 41.9%† | – |
| EBR-bench | Epoch AI Benchmarking Hub | 14.3% | – |
| GDP.pdf | Epoch AI Benchmarking Hub | 17.0% | – |
| Global-MMLU-Lite (AA) | Artificial Analysis | 93.2% | – |
| Instruction following | |||
| LiveBench Language | LiveBench | 85.4%† | – |
| LiveBench Instruction Following | LiveBench | 79.1%† | – |
| IFBench | Artificial Analysis | 77.1% | 75.9% |
| MultiChallenge | Scale AI SEAL | 71.4% | 75.5% |
| TutorBench | Scale AI SEAL | 53.0% | 68.5% |
| Arena Instruction Following | Arena (LMArena) | 1480 | 1464 |
| Human preference | |||
| Arena Text | Arena (LMArena) | 1487 | 1489 |
| EQ-Bench 4 | EQ-Bench | 1132 | – |
| Arena Creative Writing | Arena (LMArena) | 1481 | 1464 |
| Arena Multi-turn | Arena (LMArena) | 1496 | 1493 |
| Arena Longer Queries | Arena (LMArena) | 1499 | 1476 |
| Multimodal | |||
| Arena Vision | Arena (LMArena) | 1296 | 1306 |
| MMMU-Pro | Artificial Analysis | 82.4% | 80.5% |
| MMMU-Pro (official) | MMMU | 80.5%† | 80.4%† |
| MMMU-Pro (Vals) | Vals AI | 88.2%† | 87.4% |
| MortgageTax | Vals AI | 69.4%† | – |
| SAGE | Vals AI | 83.2%† | – |
| VTB | Scale AI SEAL | 29.0% | – |
| Blueprint-Bench 2 | Epoch AI Benchmarking Hub | 26.5% | – |
| Long context | |||
| AA-LCR v1.1 | Artificial Analysis | 82.0% | 78.0% |
| Arena Document | Arena (LMArena) | 1459 | 1467 |
| MLCR | Artificial Analysis | 11.1% | – |
| Composite | |||
| Epoch Capabilities Index | Epoch AI Benchmarking Hub | 156.2 | 153.1 |
| LiveBench 1.0 | LiveBench | 77.0%† | – |
| AA Intelligence Index v4.3.2 | Artificial Analysis | 29.7 | 31.3 |
| AA Progress Index v1 | Artificial Analysis | 296.1 | 289.1 |
| Vals Science Index | Vals AI | 27.4%† | – |
| Vals Multimodal Index | Vals AI | 56.1%† | – |
| Safety & honesty | |||
| FORTRESS | Scale AI SEAL | 29.8% | 20.2% |
| MASK | Scale AI SEAL | 42.4% | – |
Data as of 2026-11-06. Best configuration of each model; every score links to its source on the model pages.
What changed
- Muse Spark: BenchLeader Index 64.6 → 64.1.
- Gemini 3.1 Pro: 1 score revision.
- Gemini 3.1 Pro: 2 score revisions.
- Muse Spark: 1 score revision.
- Gemini 3.1 Pro: BenchLeader Index 61.6 → 60.5.
- Gemini 3.1 Pro: 2 new results (AA Progress Index v1, HLE-Diamond (AA)).
- Muse Spark: 1 new result (AA Progress Index v1).
- Gemini 3.1 Pro: BenchLeader Index 59.3 → 61.6.
- Gemini 3.1 Pro: Gemini 3.1 Pro cost per task moved from $1.30/M to $1.44/M (11% up).
- Gemini 3.1 Pro: 2 score revisions.
Related comparisons
- Gemini 3.1 Pro vs Claude Opus 5.5
- Gemini 3.1 Pro vs Gemini 4 Carbon
- Gemini 3.1 Pro vs GPT-6 Astra
- Muse Spark vs Claude Opus 5.5
- Muse Spark vs Gemini 4 Carbon
- Muse Spark vs GPT-6 Astra
- Muse Spark vs GPT-6.1 Sol
- Gemini 3.1 Pro vs Claude Fable 5.1
Rankings: Best LLM overall · Best LLM for agents · Best cheap LLMs
Gemini 3.1 Pro vs Muse Spark: questions
- Is Gemini 3.1 Pro better than Muse Spark?
- Gemini 3.1 Pro and Muse Spark are neck and neck on quality (60.5 vs 64.1, within the ±5.3 margin). The BenchLeader Index combines every independent quality benchmark; the two are within the margin of error as of 6 Nov 2026, so check the category scores for your use.
- Is Gemini 3.1 Pro or Muse Spark better for coding?
- Muse Spark scores higher in coding (65 vs 57 on the category index, where 50 is average). On SciCode, the top shared coding benchmark, it is 58.7% for Gemini 3.1 Pro against 51.5% for Muse Spark.
- Is Gemini 3.1 Pro or Muse Spark better for agents and tool use?
- Muse Spark scores higher in agentic tasks (65 vs 57 on the category index, where 50 is average). On Terminal-Bench Hard, the top shared agents & tools benchmark, it is 53.8% for Gemini 3.1 Pro against 45.5% for Muse Spark.
- Which is better for long documents?
- Gemini 3.1 Pro scores higher on long-context benchmarks (66 vs 64). Gemini 3.1 Pro accepts more context: 1,048,576 against 262,144 tokens for Muse Spark.
- Do independent benchmarks agree on Gemini 3.1 Pro vs Muse Spark?
- Across the 33 benchmarks both have, Gemini 3.1 Pro leads 24, Muse Spark leads 9. Every overall index that rates both puts Muse Spark ahead.
More questions (1)
- When was Muse Spark released?
- Muse Spark was released on 8 Apr 2026.