Gemini 3.6 Flash vs Muse Spark 1.2
Full profiles: Gemini 3.6 Flash (Google) and Muse Spark 1.2 (Meta): every benchmark, price and rank.
Updated · Data from 9 sources
As of 2 Nov 2026, Gemini 3.6 Flash and Muse Spark 1.2 are neck and neck on quality (61.3 vs 61.6 on the BenchLeader Index, inside the ±5.8 margin), and Gemini 3.6 Flash is 1.3× cheaper at $1.50 vs $2.00 per 1M tokens blended. Choose Gemini 3.6 Flash for agents & tools (59 vs 55) and coding (57 vs 55); choose Muse Spark 1.2 for maths (67 vs 59), reasoning (68 vs 64), and human preference (69 vs 67), where it also streams ~1.5× faster. Gemini 3.6 Flash's current price runs until 31 Dec 2026, then $3.00.
Key facts (as of 2 Nov 2026)
- Gemini 3.6 Flash scores 61.3 on the BenchLeader Index and Muse Spark 1.2 scores 61.6 (BenchLeader, 2 Nov 2026).
- Gemini 3.6 Flash costs $1.50 per 1M tokens blended until 31 Dec 2026, then $3.00 (https://ai.google.dev/gemini-api/docs/pricing, checked 20 Oct 2026).
- Muse Spark 1.2 costs $2.00 per 1M tokens blended (list price, 2 Nov 2026).
- Muse Spark 1.2 streams ~1.5× faster than Gemini 3.6 Flash (Artificial Analysis, 7-day median, 2 Nov 2026).
- Of the 59 benchmarks both models have, Muse Spark 1.2 leads 34 (BenchLeader, 2 Nov 2026).
- On Arena WebDev, Gemini 3.6 Flash is rated 1538 and Muse Spark 1.2 1534 (LMArena, as published by 2 Nov 2026).
Gemini 3.6 Flash benchmarks 
Muse Spark 1.2 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 | Gemini 3.6 Flash | 57 vs 55 category score; Arena WebDev 1538 vs 1534 | Medium |
| Agents & tool use | Gemini 3.6 Flash | 59 vs 55 category score; Terminal-Bench 2.1 (Vals) 73.8% vs 69.7% | Medium |
| Maths | Muse Spark 1.2 | 67 vs 59 category score; Arena Maths 1492 vs 1504 | High |
| Reasoning | Muse Spark 1.2 | 68 vs 64 category score; Arena Hard Prompts 1513 vs 1501 | Medium |
| Knowledge | Either | 64 vs 65 category score; SimpleQA Verified 66.2% vs 60.3% | Either |
| Long documents | Either | 65 vs 64 category score; AA-LCR v1.1 80.0% vs 79.0% | Either |
| Images and other media | Either | 65 vs 65 category score; Arena Vision 1299 vs 1304 | Either |
| Chat quality (human preference) | Muse Spark 1.2 | 69 vs 67 on Arena votes | Medium |
| Lowest cost | Gemini 3.6 Flash (until 31 Dec 2026) | $1.50 vs $2.00 per 1M blended; agentic coding session $0.060 vs $0.092; from 1 Jan 2027: $3.00 vs $2.00 | – |
| Fastest first answer | Either | 16.4 s (Gemini 3.6 Flash (high)) vs 17.5 s (Muse Spark 1.2 (xhigh)) to the first answer token | – |
| High volume on a budget | Neither flagship | A smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs | – |
Key differences
- Gemini 3.6 Flash (high) and Muse Spark 1.2 (xhigh) are neck and neck on quality (61.3 vs 61.6, within the ±5.8 margin).
- Gemini 3.6 Flash (high) is stronger in agents & tools, coding, long context.
- Muse Spark 1.2 (xhigh) is stronger in human preference, knowledge, maths, reasoning.
- Gemini 3.6 Flash (high) is 1.3× cheaper ($1.50 vs $2.00 per 1M blended) until 31 Dec 2026, then $3.00 vs $2.00 from 1 Jan 2027.
- Muse Spark 1.2 (xhigh) streams 1.5× faster (296 vs 197 tokens per second).
At a glance
| Spec | Gemini 3.6 Flash | Muse Spark 1.2 |
|---|---|---|
| Developer | Meta | |
| Released | 21 Jul 2026 | 5 Aug 2026 |
| Context window (tokens) | 1,048,576 | 1,048,576 |
| Compared at | Gemini 3.6 Flash (high) | Muse Spark 1.2 (xhigh) |
| Open weights | No | No |
| Price in / out / cached input per 1M | $0.750 / $3.75 / $0.075 | $1.25 / $4.25 / $0.150 |
| Announced price change | $1.50 / $7.50 / $0.150 from 1 Jan 2027 | None announced |
Limits, availability and announced prices from the vendors' own pages: ai.google.dev pricing (checked 20 Oct 2026). Everything else is from the day's data.
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 3.6 Flash (high) | Muse Spark 1.2 (xhigh) | Leader |
|---|---|---|---|
| Reasoning | 64.0 | 68.1 | Muse Spark 1.2 |
| Coding | 56.9 | 55.0 | Gemini 3.6 Flash |
| Agents & tools | 59.3 | 55.4 | Gemini 3.6 Flash |
| Maths | 59.4 | 67.3 | Muse Spark 1.2 |
| Knowledge | 63.8 | 64.7 | level |
| Human preference | 67.4 | 68.5 | Muse Spark 1.2 |
| Multimodal | 65.2 | 65.2 | level |
| Long context | 64.8 | 64.3 | level |
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.
Not drawn, measured for fewer than two of these models: Instruction following.
Where sources disagree
Across the 59 benchmarks both have, Gemini 3.6 Flash leads 23, Muse Spark 1.2 leads 34 and 2 are tied. Every overall index that rates both puts Muse Spark 1.2 ahead.
| Overall index | Gemini 3.6 Flash (high) | Muse Spark 1.2 (xhigh) | Leader |
|---|---|---|---|
| BenchLeader Index | 61.3 | 61.6 | neck and neck |
| Arena Text | 1483 | 1492 | Muse Spark 1.2 |
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 3.6 Flash (high) | Muse Spark 1.2 (xhigh) |
|---|---|---|
| Blended price per 1M tokens, until 31 Dec 2026 | $1.50 | $2.00 |
| Input / output per 1M | $0.750 / $3.75 | $1.25 / $4.25 |
| Blended price per 1M from 1 Jan 2027 | $3.00 | $2.00 |
| Input / output per 1M from 1 Jan 2027 | $1.50 / $7.50 | $1.25 / $4.25 |
| Cached input per 1M | $0.075 | $0.150 |
| Price per task: chat reply | $0.0014 | $0.0018 |
| Price per task: summarize a 30-page report | $0.011 | $0.018 |
| Price per task: code edit | $0.010 | $0.014 |
| Price per task: agentic coding session | $0.060 | $0.092 |
| Price per task: structured extraction | $0.0023 | $0.0034 |
| Agentic coding session, input mostly cached | $0.030 | $0.043 |
| Output speed | 197 tok/s | 296 tok/s |
| Time to first answer | 16.4 s | 17.5 s |
| Time to a complete answer | 18.9 s | 19.2 s |
Each model is compared at its best-ranked reasoning-effort setting. Gemini 3.6 Flash (high) scores 61.3 and takes 16 s to its first answer token; Gemini 3.6 Flash scores 49.5 in 1 s. Muse Spark 1.2 (xhigh) scores 61.6 and takes 18 s to its first answer token. 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
95 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 3.6 Flash (high) | Muse Spark 1.2 (xhigh) |
|---|---|---|---|
| BenchLeader Index | BenchLeader | 61.3 | 61.6 |
| Agents & tools score | BenchLeader | 59.3 | 55.4 |
| Coding score | BenchLeader | 56.9 | 55.0 |
| Human preference score | BenchLeader | 67.4 | 68.5 |
| Knowledge score | BenchLeader | 63.8 | 64.7 |
| Long context score | BenchLeader | 64.8 | 64.3 |
| Maths score | BenchLeader | 59.4 | 67.3 |
| Multimodal score | BenchLeader | 65.2 | 65.2 |
| Reasoning score | BenchLeader | 64.0 | 68.1 |
| Blended price $/M | BenchLeader | $1.50 | $2.00 |
| Output speed | BenchLeader | 197 tok/s | 296 tok/s |
| Time to first answer | BenchLeader | 16.4 s | 17.5 s |
| Context window | BenchLeader | 1,048,576 tokens | 1,048,576 tokens |
| Reasoning | |||
| GPQA Diamond | Epoch AI Benchmarking Hub | 94.1% | – |
| SimpleBench | SimpleBench | – | 74.5%† |
| Arena Hard Prompts | Arena (LMArena) | 1501 | 1513 |
| LiveBench 1.0 Reasoning | LiveBench | 85.2% | 90.0% |
| GPQA Diamond (AA) | Artificial Analysis | 92.8% | 90.4% |
| Humanity's Last Exam (AA) | Artificial Analysis | 40.8% | 45.5% |
| GPQA Diamond (Vals) | Vals AI | 93.4% | – |
| ARC-AGI-1 | ARC Prize | 91.2% | – |
| ARC-AGI-2 | ARC Prize | 60.4% | – |
| CritPt | Artificial Analysis | 10.6% | 17.7% |
| Chess Puzzles | Epoch AI Benchmarking Hub | 40.0% | 27.0% |
| Mystery Game Puzzles | Epoch AI Benchmarking Hub | 30.0% | – |
| LMCA | Epoch AI Benchmarking Hub | 44.9% | 48.4% |
| DTBench | Epoch AI Benchmarking Hub | 95.5% | 94.7% |
| ForecastBench | ForecastBench | 65.9%† | – |
| Coding | |||
| SciCode | SciCode | 53.4% | 57.4% |
| WeirdML | WeirdML | 56.1% | 60.3% |
| FrontierCode | Cognition | 34.4%† | – |
| Arena Coding | Arena (LMArena) | 1518 | 1543 |
| Arena WebDev | Arena (LMArena) | 1538 | 1534 |
| LiveBench 1.0 Coding | LiveBench | 77.9% | 77.5% |
| LiveCodeBench | Vals AI | 88.1% | – |
| IOI | Vals AI | 35.1% | 21.8% |
| SWE-bench (Vals) | Vals AI | 79.6% | 86.6% |
| Code Migration | Vals AI | 30.9% | 29.9% |
Show all 95 benchmarks
| Metric | Publisher | Gemini 3.6 Flash (high) | Muse Spark 1.2 (xhigh) |
|---|---|---|---|
| Coding | |||
| ProgramBench | Vals AI | 0.0% | 0.5% |
| Vibe Code Bench v1.1 | Vals AI | 64.0% | 79.1% |
| DeepSWE v1.1 | Epoch AI Benchmarking Hub | 46.7% | 54.9% |
| ALE-Bench | Epoch AI Benchmarking Hub | 715.5 | – |
| FrontierSWE | Epoch AI Benchmarking Hub | – | 12.0% |
| Terminal-Bench 4.0 (AA) | Artificial Analysis | 7.1% | 7.1% |
| Terminal-Bench 2.1 (AA) | Artificial Analysis | 77.5% | 80.2% |
| Startrise Frontend | Startrise | – | 70.3%† |
| GameBench 2 Overall | DuelLab GameBench 2 | 52.5† | 53.7† |
| GameBench 2 by reasoning setting | DuelLab GameBench 2 | 64.6 | 59.7 |
| Agents & tools | |||
| APEX-Agents | Mercor | 46.9%† | 45.4%† |
| Arena Agent | Arena (LMArena) | – | -7.5 |
| LiveBench Agentic Coding | LiveBench | 43.4% | 57.6% |
| Terminal-Bench 2.1 (Vals) | Vals AI | 73.8% | 69.7% |
| τ³-Banking (AA) | Artificial Analysis | 29.9% | 34.9% |
| CyberBench | Vals AI | 45.4% | 69.5% |
| Finance Agent v2 | Vals AI | 56.3% | 60.6% |
| Harvey's Legal Agent Benchmark v1 | Vals AI | 3.3% | 25.4% |
| Legal Research Bench | Vals AI | 25.0% | 43.8% |
| SkillsBench | Vals AI | – | 53.0% |
| SREBench | Vals AI | – | 0.0% |
| Tax Agent Bench | Vals AI | 18.0% | 32.8% |
| Terminal-Bench 4.0 (Vals) | Vals AI | – | 6.1% |
| Terminal-Bench Science | Vals AI | 4.3% | – |
| AutomationBench | Artificial Analysis | 53.0% | 40.6% |
| GDP.pdf (AA) | Artificial Analysis | 17.4% | 17.4% |
| EnterpriseOps-Gym | Artificial Analysis | – | 47.3% |
| AA-Briefcase v1.1 | Artificial Analysis | – | 1325 |
| FACTS Search | 68.2%† | – | |
| Maths | |||
| FrontierMath Tiers 1–3 | Epoch AI Benchmarking Hub | 59.0% | – |
| FrontierMath Tier 4 | Epoch AI Benchmarking Hub | 21.9% | – |
| OTIS Mock AIME | Epoch AI Benchmarking Hub | 94.2% | – |
| ProofBench | Vals AI | 36.0%† | 43.0%† |
| LiveBench Mathematics | LiveBench | 86.4% | 91.2% |
| AIME 2026 | MathArena | 96.7%† | – |
| HMMT February 2026 | MathArena | 89.4%† | – |
| MathArena Apex | MathArena | 26.0%† | – |
| Arena Maths | Arena (LMArena) | 1504 | 1492 |
| Knowledge | |||
| SimpleQA Verified | Epoch AI Benchmarking Hub | 66.2% | 60.3% |
| LiveBench 1.0 Data Analysis | LiveBench | 63.0% | 76.5% |
| MMLU-Pro | Vals AI | 89.3% | 88.3% |
| LegalBench | Vals AI | 86.7% | 85.3% |
| CorpFin | Vals AI | 63.3% | 70.9% |
| TaxEval | Vals AI | 74.9% | 80.4% |
| BioMysteryBench | Vals AI | 58.5% | 64.8% |
| Excel Modeling Benchmark | Vals AI | 65.4% | 57.0% |
| MedCode | Vals AI | 53.1% | 49.4% |
| MedScribe | Vals AI | 79.7% | 90.1% |
| Public Benefits Bench v1.1 | Vals AI | 56.6% | 68.5% |
| GDP.pdf | Epoch AI Benchmarking Hub | 14.0% | 16.0% |
| Instruction following | |||
| LiveBench Language | LiveBench | 83.9% | 78.6% |
| LiveBench Instruction Following | LiveBench | 75.4% | 74.3% |
| Arena Instruction Following | Arena (LMArena) | 1473 | 1475 |
| Human preference | |||
| Arena Text | Arena (LMArena) | 1483 | 1492 |
| Arena Creative Writing | Arena (LMArena) | 1471 | 1441 |
| Arena Multi-turn | Arena (LMArena) | 1488 | 1508 |
| Arena Longer Queries | Arena (LMArena) | 1485 | 1489 |
| Multimodal | |||
| Arena Vision | Arena (LMArena) | 1299 | 1304 |
| MMMU-Pro | Artificial Analysis | 83.2% | – |
| MMMU-Pro (Vals) | Vals AI | 88.4% | 86.1% |
| MortgageTax | Vals AI | 67.6% | 65.4% |
| Blueprint-Bench 2 | Epoch AI Benchmarking Hub | 31.2%† | – |
| Long context | |||
| AA-LCR v1.1 | Artificial Analysis | 80.0% | 79.0% |
| Arena Document | Arena (LMArena) | 1453 | – |
| MLCR | Artificial Analysis | 10.0% | 7.2% |
| Composite | |||
| Epoch Capabilities Index | Epoch AI Benchmarking Hub | 154.7† | 155.2† |
| LiveBench 1.0 | LiveBench | 73.6% | 78.0% |
| AA Intelligence Index v4.3.2 | Artificial Analysis | 34.0 | 39.6 |
| AA Progress Index v1 | Artificial Analysis | 305.7 | 309.1 |
| Vals Multimodal Index | Vals AI | 65.1% | 69.8%† |
Data as of 2026-11-02. Best configuration of each model; every score links to its source on the model pages.
DuelLab GameBench 2, release gb2-20261026-cb69502a12d9 (https://benchmarks.duellab.org/releases/gb2-20261026-cb69502a12d9/), CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Scores mapped to BenchLeader model names; not endorsed by DuelLab.
What changed
- Gemini 3.6 Flash: list price moves to $1.50 / $7.50 per 1M input / output tokens (announced).
- Muse Spark 1.2 (xhigh): 1 score revision.
- Gemini 3.6 Flash (high): BenchLeader Index 61.8 → 61.2.
- Muse Spark 1.2 (xhigh): BenchLeader Index 62.2 → 61.6.
- Gemini 3.6 Flash (high): 1 new result (AA Progress Index v1).
- Muse Spark 1.2 (xhigh): 1 new result (AA Progress Index v1).
- Muse Spark 1.2 (xhigh): cost per task moved from $1.29/M to $0.81/M (37% down).
- Muse Spark 1.2 (xhigh): cost per task moved from $0.97/M to $1.29/M (32% up).
- Muse Spark 1.2 (xhigh): 1 new result (AA-Briefcase v1.1).
- Gemini 3.6 Flash (high): 1 new result (Harvey's Legal Agent Benchmark v1).
Related comparisons
- Gemini 3.6 Flash vs Claude Opus 5.5
- Gemini 3.6 Flash vs Gemini 4 Carbon
- Gemini 3.6 Flash vs GPT-6 Astra
- Muse Spark 1.2 vs Claude Opus 5.5
- Muse Spark 1.2 vs Gemini 4 Carbon
- Muse Spark 1.2 vs GPT-6 Astra
- Muse Spark 1.2 vs GPT-6.1 Sol
- Gemini 3.5 Flash vs Muse Spark 1.2
Rankings: Best LLM overall · Best LLM for agents · Best cheap LLMs
Gemini 3.6 Flash vs Muse Spark 1.2: questions
- Is Gemini 3.6 Flash better than Muse Spark 1.2?
- Gemini 3.6 Flash (high) and Muse Spark 1.2 (xhigh) are neck and neck on quality (61.3 vs 61.6, within the ±5.8 margin). The BenchLeader Index combines every independent quality benchmark; the two are within the margin of error as of 2 Nov 2026, so check the category scores for your use.
- Gemini 3.6 Flash vs Muse Spark 1.2: which is cheaper, and will that change?
- Gemini 3.6 Flash is cheaper today: $1.50 against $2.00 per million tokens, blended at three input tokens per output token. Gemini 3.6 Flash's current price runs until 31 Dec 2026, then $3.00.
- Is Gemini 3.6 Flash or Muse Spark 1.2 better for coding?
- Gemini 3.6 Flash scores higher in coding (57 vs 55 on the category index, where 50 is average). On Arena WebDev, the top shared coding benchmark, it is 1538 for Gemini 3.6 Flash against 1534 for Muse Spark 1.2.
- Which is faster, Gemini 3.6 Flash or Muse Spark 1.2?
- Muse Spark 1.2 streams ~1.5× faster (296 vs 197 output tokens per second, Artificial Analysis 7-day median).
- Is Gemini 3.6 Flash or Muse Spark 1.2 better for agents and tool use?
- Gemini 3.6 Flash scores higher in agentic tasks (59 vs 55 on the category index, where 50 is average). On τ³-Banking (AA), the top shared agents & tools benchmark, it is 29.9% for Gemini 3.6 Flash against 34.9% for Muse Spark 1.2.
More questions (4)
- Which is better for long documents?
- They are level on long-context benchmarks (65 vs 64). Both accept 1,048,576 tokens of context.
- Which has lower latency?
- Gemini 3.6 Flash answers first: 16.4 s against 17.5 s to the first answer token, as Artificial Analysis measures it. Reasoning effort drives most of this: a higher setting thinks longer before it answers.
- Do independent benchmarks agree on Gemini 3.6 Flash vs Muse Spark 1.2?
- Across the 59 benchmarks both have, Gemini 3.6 Flash leads 23, Muse Spark 1.2 leads 34 and 2 are tied. Every overall index that rates both puts Muse Spark 1.2 ahead.
- When was Muse Spark 1.2 released?
- Muse Spark 1.2 was released on 5 Aug 2026.