Muse Spark 1.3 Max vs GPT-5.6 Sol: Benchmark Comparison
Muse Spark 1.3 Max has the higher score on 9 of 18 shared benchmarks; GPT-5.6 Sol leads on 9.
The largest observed score gap is 34.61 pts on IOI , where GPT-5.6 Sol leads.
Reported ±1 standard-error ranges overlap on 9 of 16 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 1.3 Max | GPT-5.6 Sol | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals Index | 58.16% ±1.19 | 58.01% ±1.02 | 0.16 pts | Reported ±1 SE ranges overlap |
| Vals RSI Index | 19.64% | 23.88% | 4.24 pts | Uncertainty comparison unavailable |
| Harvey's Legal Agent Benchmark | 23.75% ±3.55 | 2.50% ±0.83 | 21.25 pts | Reported ±1 SE ranges do not overlap |
| Legal Research Bench | 55.29% ±3.46 | 48.08% ±3.47 | 7.21 pts | Reported ±1 SE ranges do not overlap |
| EMB | 67.43% ±3.06 | 72.34% ±2.37 | 4.91 pts | Reported ±1 SE ranges overlap |
| Finance Agent (v2) | 59.96% ±2.06 | 53.76% ±0.85 | 6.20 pts | Reported ±1 SE ranges do not overlap |
| Tax Agent Bench | 72.44% ±2.88 | 67.95% ±3.12 | 4.49 pts | Reported ±1 SE ranges overlap |
| ProofBench v1.1 | 58.00% ±4.96 | 83.00% ±3.77 | 25.00 pts | Reported ±1 SE ranges do not overlap |
| MysteryMechanism | 36.04% ±3.23 | 33.33% ±3.17 | 2.70 pts | Reported ±1 SE ranges overlap |
| Terminal-Bench Science | 10.00% ±3.61 | 20.00% ±4.82 | 10.00 pts | Reported ±1 SE ranges do not overlap |
| Code Migration | 47.41% ±4.26 | 52.92% ±4.35 | 5.50 pts | Reported ±1 SE ranges overlap |
| IOI | 56.56% ±2.52 | 91.17% ±4.51 | 34.61 pts | Reported ±1 SE ranges do not overlap |
| ProgramBench | 2.50% ±1.11 | 1.50% ±0.86 | 1.00 pts | Reported ±1 SE ranges overlap |
| Terminal-Bench 4.0 | 24.75% ±0.51 | 37.88% ±0.00 | 13.13 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench 1-100 | 20.46% ±3.85 | 20.01% ±3.63 | 0.45 pts | Reported ±1 SE ranges overlap |
| Vibe Code Bench v1.1 | 85.86% ±2.51 | 80.50% ±3.72 | 5.36 pts | Reported ±1 SE ranges overlap |
| CyberBench v1.1 | 72.74% ±5.67 | 76.31% ±5.18 | 3.57 pts | Reported ±1 SE ranges overlap |
| CUA-bench | 5.83% | 8.33% | 2.50 pts | Uncertainty comparison unavailable |
Performance by category
| Category | Muse Spark 1.3 Max average | GPT-5.6 Sol average |
|---|---|---|
| Legal | 39.52% | 25.29% |
| Finance | 66.61% | 64.68% |
| Math | 58.00% | 83.00% |
| Science | 23.02% | 26.67% |
| Coding | 39.59% | 47.33% |
| Cyber | 72.74% | 76.31% |
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 1.3 Max cost | GPT-5.6 Sol cost | Muse Spark 1.3 Max latency | GPT-5.6 Sol latency |
|---|---|---|---|---|
| Vals Index | $3.79 | $14.24 | 23m33s | 37m23s |
| Vals RSI Index | $48.82 | $347.10 | 90h00m | 90h00m |
| Harvey's Legal Agent Benchmark | $2.26 | $10.37 | 14m35s | 22m50s |
| Legal Research Bench | $0.59 | $21.61 | 5m17s | 1h17m |
| EMB | $2.55 | $6.01 | 13m50s | 9m50s |
| Finance Agent (v2) | $0.76 | $1.25 | 3m19s | 19m25s |
| Tax Agent Bench | $0.39 | $7.51 | 4m09s | 44m41s |
| ProofBench v1.1 | $0.48 | $1.44 | 9m29s | 8m40s |
| MysteryMechanism | $0.85 | $0.96 | 7m32s | 8m43s |
| Terminal-Bench Science | $6.09 | $7.32 | 1h24m | 1h44m |
| Code Migration | $14.21 | $24.54 | 1h24m | 59m12s |
| IOI | $1.73 | $7.98 | 28m58s | 1h02m |
| ProgramBench | $54.29 | $15.29 | 5h28m | 29m24s |
| Terminal-Bench 4.0 | $6.65 | $7.98 | 52m56s | 33m46s |
| Vibe Code Bench 1-100 | $5.60 | $47.43 | 20m02s | 2h37m |
| Vibe Code Bench v1.1 | $2.54 | $33.40 | 16m18s | 33m53s |
| CyberBench v1.1 | $3.34 | $3.05 | 16m18s | 10m47s |
| CUA-bench | $143.48 | $206.99 | 36.39s | 18.80s |
Results available only for Muse Spark 1.3 Max
None.
Results available only for GPT-5.6 Sol
- LegalBench
- MortgageTax
- TaxEval v2
- MedCode
- MedScribe
- BioMysteryBench
- GPQA Diamond
- MMLU Pro
- MMMU Pro
- SAGE
- LiveCodeBench
- SkillsBench
- SWE-bench
- SRE Bench
- Public Benefits Bench v1.1
- Time Horizon Index: KSP