Gemini 3.6 Flash vs Muse Spark 1.2: Benchmark Comparison
Gemini 3.6 Flash has the higher score on 9 of 21 shared benchmarks; Muse Spark 1.2 leads on 12.
The largest observed score gap is 24.17 pts on CyberBench v1.1 , where Muse Spark 1.2 leads.
Reported ±1 standard-error ranges overlap on 4 of 21 shared benchmarks with comparable uncertainty data. This is not a pairwise statistical significance test.
Shared benchmark results
Scores and reported standard errors come from the latest visible Vals benchmark version available for both models.
| Benchmark | Gemini 3.6 Flash | Muse Spark 1.2 | Gap | Reported uncertainty |
|---|---|---|---|---|
| Harvey's Legal Agent Benchmark | 3.33% ±1.17 | 25.42% ±3.67 | 22.08 pts | Reported ±1 SE ranges do not overlap |
| Legal Research Bench | 25.00% ±3.01 | 43.75% ±3.45 | 18.75 pts | Reported ±1 SE ranges do not overlap |
| LegalBench | 86.70% ±0.41 | 85.26% ±0.45 | 1.44 pts | Reported ±1 SE ranges do not overlap |
| EMB | 65.41% ±2.65 | 56.98% ±3.13 | 8.43 pts | Reported ±1 SE ranges do not overlap |
| Finance Agent (v2) | 56.30% ±0.18 | 60.60% ±0.28 | 4.30 pts | Reported ±1 SE ranges do not overlap |
| MortgageTax | 67.61% ±0.92 | 65.42% ±0.93 | 2.19 pts | Reported ±1 SE ranges do not overlap |
| Tax Agent Bench | 49.90% ±3.30 | 56.86% ±2.32 | 6.96 pts | Reported ±1 SE ranges do not overlap |
| TaxEval v2 | 74.86% ±0.85 | 80.38% ±0.76 | 5.52 pts | Reported ±1 SE ranges do not overlap |
| MedCode | 53.15% ±2.16 | 49.35% ±2.19 | 3.81 pts | Reported ±1 SE ranges overlap |
| MedScribe | 79.66% ±1.86 | 90.06% ±1.96 | 10.40 pts | Reported ±1 SE ranges do not overlap |
| BioMysteryBench | 58.52% ±1.61 | 64.81% ±1.33 | 6.30 pts | Reported ±1 SE ranges do not overlap |
| MMLU Pro | 89.28% ±0.30 | 88.28% ±0.32 | 1.00 pts | Reported ±1 SE ranges do not overlap |
| MMMU Pro | 88.38% ±0.77 | 86.13% ±0.83 | 2.26 pts | Reported ±1 SE ranges do not overlap |
| SAGE | 48.69% ±3.38 | 47.66% ±3.44 | 1.03 pts | Reported ±1 SE ranges overlap |
| Code Migration | 30.93% ±4.07 | 29.95% ±4.02 | 0.97 pts | Reported ±1 SE ranges overlap |
| IOI | 35.06% ±5.70 | 21.78% ±0.87 | 13.28 pts | Reported ±1 SE ranges do not overlap |
| ProgramBench | 0.00% ±0.00 | 0.50% ±0.50 | 0.50 pts | Reported ±1 SE ranges overlap |
| SWE-bench | 79.60% ±1.80 | 86.60% ±1.52 | 7.00 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 64.00% ±4.29 | 79.10% ±3.31 | 15.09 pts | Reported ±1 SE ranges do not overlap |
| CyberBench v1.1 | 45.36% ±3.75 | 69.52% ±5.56 | 24.17 pts | Reported ±1 SE ranges do not overlap |
| Public Benefits Bench v1.1 | 56.56% ±1.29 | 68.47% ±1.21 | 11.91 pts | Reported ±1 SE ranges do not overlap |
Performance by category
| Category | Gemini 3.6 Flash average | Muse Spark 1.2 average |
|---|---|---|
| Legal | 38.35% | 51.48% |
| Finance | 62.81% | 64.05% |
| Healthcare | 66.41% | 69.70% |
| Science | 58.52% | 64.81% |
| Academic | 88.83% | 87.20% |
| Education | 48.69% | 47.66% |
| Coding | 41.92% | 43.59% |
| Cyber | 45.36% | 69.52% |
| Social Mobility | 56.56% | 68.47% |
Cost and latency
Cost per test appears only where the benchmark reports it for both models. Latency is the measured completion time for that benchmark.
| Benchmark | Gemini 3.6 Flash cost | Muse Spark 1.2 cost | Gemini 3.6 Flash latency | Muse Spark 1.2 latency |
|---|---|---|---|---|
| Harvey's Legal Agent Benchmark | $1.90 | $2.09 | 16m24s | 24m16s |
| Legal Research Bench | $0.47 | $0.51 | 2m48s | 5m07s |
| LegalBench | N/A | N/A | 2.96s | 30.10s |
| EMB | $3.89 | $2.31 | 12m06s | 17m07s |
| Finance Agent (v2) | $1.40 | $0.77 | 3m45s | 5m04s |
| MortgageTax | N/A | N/A | 9.24s | 44.62s |
| Tax Agent Bench | $0.36 | $0.27 | 105.32s | 119.91s |
| TaxEval v2 | N/A | N/A | 12.34s | 48.31s |
| MedCode | N/A | N/A | 16.68s | 60.17s |
| MedScribe | N/A | N/A | 33.55s | 61.46s |
| BioMysteryBench | $1.25 | $0.99 | 12m42s | 23m37s |
| MMLU Pro | N/A | N/A | 7.30s | 44.23s |
| MMMU Pro | N/A | N/A | 11.31s | 73.84s |
| SAGE | N/A | N/A | 21.96s | 29.07s |
| Code Migration | $8.93 | $3.78 | 1h30m | 31m07s |
| IOI | $8.67 | $2.64 | 48m23s | 25m35s |
| ProgramBench | $5.92 | $2.47 | 47m33s | 26m38s |
| SWE-bench | $1.19 | $0.55 | 13m08s | 8m07s |
| Vibe Code Bench v1.1 | $3.04 | $1.53 | 25m23s | 19m55s |
| CyberBench v1.1 | $1.31 | $2.15 | 11m00s | 20m21s |
| Public Benefits Bench v1.1 | $0.39 | $0.59 | 2m02s | 8m55s |
Results available only for Gemini 3.6 Flash
- Terminal-Bench Science
- GPQA Diamond
- LiveCodeBench
Results available only for Muse Spark 1.2
- Vals Index
- ProofBench v1.1
- SkillsBench
- Terminal-Bench 4.0