Muse Spark 1.2 vs GPT 5.5: Benchmark Comparison
Muse Spark 1.2 has the higher score on 10 of 19 shared benchmarks; GPT 5.5 leads on 8.
The largest observed score gap is 21.67 pts on Harvey's Legal Agent Benchmark , where Muse Spark 1.2 leads.
Reported ±1 standard-error ranges overlap on 7 of 19 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 | Muse Spark 1.2 | GPT 5.5 | Gap | Reported uncertainty |
|---|---|---|---|---|
| Harvey's Legal Agent Benchmark | 25.42% ±3.67 | 3.75% ±1.17 | 21.67 pts | Reported ±1 SE ranges do not overlap |
| Legal Research Bench | 43.75% ±3.45 | 40.38% ±3.41 | 3.37 pts | Reported ±1 SE ranges overlap |
| LegalBench | 85.26% ±0.45 | 86.52% ±0.41 | 1.25 pts | Reported ±1 SE ranges do not overlap |
| EMB | 56.98% ±3.13 | 64.54% ±2.87 | 7.57 pts | Reported ±1 SE ranges do not overlap |
| Finance Agent (v2) | 60.60% ±0.28 | 51.76% ±0.55 | 8.84 pts | Reported ±1 SE ranges do not overlap |
| MortgageTax | 65.42% ±0.93 | 68.76% ±0.91 | 3.34 pts | Reported ±1 SE ranges do not overlap |
| Tax Agent Bench | 56.86% ±2.32 | 60.46% ±3.22 | 3.60 pts | Reported ±1 SE ranges overlap |
| TaxEval v2 | 80.38% ±0.76 | 74.98% ±0.85 | 5.40 pts | Reported ±1 SE ranges do not overlap |
| MedCode | 49.35% ±2.19 | 49.10% ±2.19 | 0.25 pts | Reported ±1 SE ranges overlap |
| MedScribe | 90.06% ±1.96 | 86.87% ±1.93 | 3.19 pts | Reported ±1 SE ranges overlap |
| MMLU Pro | 88.28% ±0.32 | 88.14% ±0.32 | 0.13 pts | Reported ±1 SE ranges overlap |
| MMMU Pro | 86.13% ±0.83 | 88.27% ±0.77 | 2.14 pts | Reported ±1 SE ranges do not overlap |
| SAGE | 47.66% ±3.44 | 51.53% ±3.95 | 3.87 pts | Reported ±1 SE ranges overlap |
| Code Migration | 29.95% ±4.02 | 45.16% ±4.16 | 15.21 pts | Reported ±1 SE ranges do not overlap |
| ProgramBench | 0.50% ±0.50 | 0.50% ±0.50 | 0.00 pts | Reported ±1 SE ranges overlap |
| SkillsBench | 53.04% ±4.42 | 62.21% ±4.52 | 9.17 pts | Reported ±1 SE ranges do not overlap |
| SWE-bench | 86.60% ±1.52 | 82.60% ±1.70 | 4.00 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 79.10% ±3.31 | 69.85% ±4.54 | 9.25 pts | Reported ±1 SE ranges do not overlap |
| Public Benefits Bench v1.1 | 68.47% ±1.21 | 60.89% ±1.27 | 7.58 pts | Reported ±1 SE ranges do not overlap |
Performance by category
| Category | Muse Spark 1.2 average | GPT 5.5 average |
|---|---|---|
| Legal | 51.48% | 43.55% |
| Finance | 64.05% | 64.10% |
| Healthcare | 69.70% | 67.98% |
| Academic | 87.20% | 88.21% |
| Education | 47.66% | 51.53% |
| Coding | 49.84% | 52.06% |
| Social Mobility | 68.47% | 60.89% |
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 | Muse Spark 1.2 cost | GPT 5.5 cost | Muse Spark 1.2 latency | GPT 5.5 latency |
|---|---|---|---|---|
| Harvey's Legal Agent Benchmark | $2.09 | $4.60 | 24m16s | 12m14s |
| Legal Research Bench | $0.51 | $7.40 | 5m07s | 37m45s |
| LegalBench | N/A | N/A | 30.10s | 18.14s |
| EMB | $2.31 | $3.27 | 17m07s | 15m11s |
| Finance Agent (v2) | $0.77 | $4.15 | 5m04s | 11m02s |
| MortgageTax | N/A | N/A | 44.62s | 28.24s |
| Tax Agent Bench | $0.27 | $4.07 | 119.91s | 15m52s |
| TaxEval v2 | N/A | N/A | 48.31s | 95.83s |
| MedCode | N/A | N/A | 60.17s | 2m40s |
| MedScribe | N/A | N/A | 61.46s | 2m13s |
| MMLU Pro | N/A | N/A | 44.23s | 42.11s |
| MMMU Pro | N/A | N/A | 73.84s | 54.15s |
| SAGE | N/A | N/A | 29.07s | 76.05s |
| Code Migration | $3.78 | $6.44 | 31m07s | 31m35s |
| ProgramBench | $2.47 | $6.95 | 26m38s | 22m03s |
| SkillsBench | $0.65 | $2.54 | 13m00s | 6m55s |
| SWE-bench | $0.55 | $1.36 | 8m07s | 7m06s |
| Vibe Code Bench v1.1 | $1.53 | $16.66 | 19m55s | 31m52s |
| Public Benefits Bench v1.1 | $0.59 | $3.97 | 8m55s | 39m37s |
Results available only for Muse Spark 1.2
- Vals Index
- ProofBench v1.1
- BioMysteryBench
- IOI
- Terminal-Bench 4.0
- CyberBench v1.1
Results available only for GPT 5.5
- Vals RSI Index
- GPQA Diamond
- LiveCodeBench
- SRE Bench
- Time Horizon Index: KSP