Muse Spark 1.3 Max vs GPT-6 Astra: Benchmark Comparison
Muse Spark 1.3 Max has the higher score on 4 of 16 shared benchmarks; GPT-6 Astra leads on 12.
The largest observed score gap is 53.05 pts on SRE Bench , where GPT-6 Astra leads.
Reported ±1 standard-error ranges overlap on 4 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 Max | GPT-6 Astra | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals RSI Index | 19.64% | 27.07% | 7.43 pts | Uncertainty comparison unavailable |
| Immigration | 64.93% ±1.56 | 67.69% ±1.55 | 2.76 pts | Reported ±1 SE ranges overlap |
| Legal Research Bench | 55.29% ±3.46 | 39.42% ±3.40 | 15.86 pts | Reported ±1 SE ranges do not overlap |
| Tax Agent Bench | 42.44% ±3.51 | 20.72% ±2.92 | 21.72 pts | Reported ±1 SE ranges do not overlap |
| ProofBench v1.1 | 58.00% ±4.96 | 99.00% ±1.00 | 41.00 pts | Reported ±1 SE ranges do not overlap |
| MysteryMechanism | 36.04% ±3.23 | 53.15% ±3.36 | 17.12 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench-Science | 10.00% ±3.61 | 62.86% ±5.82 | 52.86 pts | Reported ±1 SE ranges do not overlap |
| IOI | 56.56% ±2.52 | 100.00% ±0.00 | 43.44 pts | Reported ±1 SE ranges do not overlap |
| ProgramBench | 2.50% ±1.11 | 5.50% ±1.62 | 3.00 pts | Reported ±1 SE ranges do not overlap |
| SWE-bench Multimodal | 43.13% ±2.26 | 43.33% ±2.26 | 0.21 pts | Reported ±1 SE ranges overlap |
| Terminal-Bench 4.0 | 24.75% ±0.51 | 59.60% ±4.40 | 34.85 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 85.86% ±2.51 | 89.59% ±2.17 | 3.74 pts | Reported ±1 SE ranges overlap |
| CyberBench v1.1 | 72.74% ±5.67 | 41.07% ±2.56 | 31.67 pts | Reported ±1 SE ranges do not overlap |
| SRE Bench | 3.82% ±1.19 | 56.87% ±3.07 | 53.05 pts | Reported ±1 SE ranges do not overlap |
| Financial Coaching Bench | 70.29% ±1.68 | 67.53% ±1.69 | 2.75 pts | Reported ±1 SE ranges overlap |
| CUA-bench | 5.83% | 19.17% | 13.33 pts | Uncertainty comparison unavailable |
Performance by category
| Category | Muse Spark 1.3 Max average | GPT-6 Astra average |
|---|---|---|
| Legal | 60.11% | 53.56% |
| Finance | 42.44% | 20.72% |
| Math | 58.00% | 99.00% |
| Science | 23.02% | 58.00% |
| Coding | 42.56% | 59.60% |
| Cyber | 38.28% | 48.97% |
| Social Mobility | 70.29% | 67.53% |
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-6 Astra cost | Muse Spark 1.3 Max latency | GPT-6 Astra latency |
|---|---|---|---|---|
| Vals RSI Index | $48.82 | $759.51 | 90h00m | 90h00m |
| Immigration | $0.06 | $1.37 | 97.63s | 8m21s |
| Legal Research Bench | $0.59 | $10.58 | 5m17s | 24m51s |
| Tax Agent Bench | $0.39 | $5.80 | 4m09s | 15m01s |
| ProofBench v1.1 | $0.48 | $1.71 | 9m29s | 5m08s |
| MysteryMechanism | $0.85 | $1.56 | 7m32s | 8m09s |
| Terminal-Bench-Science | $6.09 | $20.80 | 1h24m | 1h49m |
| IOI | $1.73 | $6.50 | 28m58s | 31m40s |
| ProgramBench | $54.29 | $11.57 | 5h28m | 27m02s |
| SWE-bench Multimodal | $0.86 | $1.81 | 10m52s | 7m03s |
| Terminal-Bench 4.0 | $6.65 | $9.58 | 52m56s | 35m42s |
| Vibe Code Bench v1.1 | $2.54 | $38.51 | 16m18s | 39m04s |
| CyberBench v1.1 | $3.34 | $1.84 | 16m18s | 4m10s |
| SRE Bench | $25.86 | $10.24 | 1h47m | 17m51s |
| Financial Coaching Bench | $0.09 | $1.72 | 2m58s | 12m33s |
| CUA-bench | $143.48 | $1821.37 | 36.39s | 32.09s |
Results available only for Muse Spark 1.3 Max
- Access to Justice Benchmark
Results available only for GPT-6 Astra
- Vals Index
- Vals Finance Index
- Vals Legal Index
- Vals Science Index
- HLAB-Vals-1.1
- EMB v1.1
- Finance Agent v2.1
- MedCode
- MedScribe
- BioMysteryBench
- Robo-bench
- Code Migration v1.1
- Time Horizon Index: KSP