Muse Spark 1.3 vs GPT-5.6 Terra: Benchmark Comparison
Muse Spark 1.3 has the higher score on 5 of 14 shared benchmarks; GPT-5.6 Terra leads on 8.
The largest observed score gap is 43.67 pts on IOI , where GPT-5.6 Terra leads.
Reported ±1 standard-error ranges overlap on 6 of 14 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 | GPT-5.6 Terra | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals Index | 53.20% ±1.07 | 53.09% ±1.29 | 0.11 pts | Reported ±1 SE ranges overlap |
| Harvey's Legal Agent Benchmark | 22.92% ±3.42 | 0.83% ±0.83 | 22.08 pts | Reported ±1 SE ranges do not overlap |
| Legal Research Bench | 40.87% ±3.42 | 41.35% ±3.42 | 0.48 pts | Reported ±1 SE ranges overlap |
| EMB | 62.71% ±2.96 | 66.20% ±3.11 | 3.49 pts | Reported ±1 SE ranges overlap |
| Finance Agent (v2) | 58.90% ±0.44 | 54.44% ±2.07 | 4.47 pts | Reported ±1 SE ranges do not overlap |
| Tax Agent Bench | 71.93% ±2.92 | 65.20% ±3.19 | 6.73 pts | Reported ±1 SE ranges do not overlap |
| ProofBench v1.1 | 55.00% ±5.00 | 74.00% ±4.41 | 19.00 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench Science | 4.29% ±2.44 | 10.00% ±3.61 | 5.71 pts | Reported ±1 SE ranges overlap |
| Code Migration | 27.58% ±4.07 | 47.80% ±4.28 | 20.22 pts | Reported ±1 SE ranges do not overlap |
| IOI | 43.94% ±1.45 | 87.61% ±6.53 | 43.67 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 |
| Terminal-Bench 4.0 | 10.61% ±1.75 | 22.73% ±2.31 | 12.12 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 82.86% ±2.90 | 74.59% ±4.20 | 8.27 pts | Reported ±1 SE ranges do not overlap |
| CyberBench v1.1 | 69.41% ±5.76 | 72.08% ±5.41 | 2.68 pts | Reported ±1 SE ranges overlap |
Performance by category
| Category | Muse Spark 1.3 average | GPT-5.6 Terra average |
|---|---|---|
| Legal | 31.89% | 21.09% |
| Finance | 64.52% | 61.95% |
| Math | 55.00% | 74.00% |
| Science | 4.29% | 10.00% |
| Coding | 33.10% | 46.65% |
| Cyber | 69.41% | 72.08% |
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 cost | GPT-5.6 Terra cost | Muse Spark 1.3 latency | GPT-5.6 Terra latency |
|---|---|---|---|---|
| Vals Index | $3.38 | $5.78 | 32m42s | 35m39s |
| Harvey's Legal Agent Benchmark | $2.76 | $3.20 | 16m39s | 17m05s |
| Legal Research Bench | $0.65 | $7.91 | 8m51s | 1h10m |
| EMB | $3.55 | $2.19 | 22m55s | 13m33s |
| Finance Agent (v2) | $0.74 | $3.63 | 5m42s | 24m10s |
| Tax Agent Bench | $0.32 | $4.83 | 8m25s | 1h01m |
| ProofBench v1.1 | $0.43 | $1.27 | 8m54s | 12m57s |
| Terminal-Bench Science | $4.41 | $5.19 | 1h08m | 2h04m |
| Code Migration | $4.62 | $8.13 | 1h08m | 48m34s |
| IOI | $2.36 | $8.66 | 29m51s | 1h23m |
| ProgramBench | $11.56 | $5.38 | 1h09m | 33m36s |
| Terminal-Bench 4.0 | $7.04 | $5.60 | 1h59m | 31m29s |
| Vibe Code Bench v1.1 | $2.10 | $7.89 | 12m54s | 19m12s |
| CyberBench v1.1 | $4.26 | $3.31 | 24m53s | 17m04s |
Results available only for Muse Spark 1.3
None.
Results available only for GPT-5.6 Terra
- LegalBench
- MortgageTax
- TaxEval v2
- MedCode
- MedScribe
- GPQA Diamond
- MMLU Pro
- MMMU Pro
- SAGE
- LiveCodeBench
- SkillsBench
- SWE-bench
- Vibe Code Bench 1-100
- Public Benefits Bench v1.1