Claude Opus 5 vs GPT-5.6 Sol: Benchmark Comparison
Claude Opus 5 has the higher score on 28 of 34 shared benchmarks; GPT-5.6 Sol leads on 6.
The largest observed score gap is 19.60 pts on MedCode , where Claude Opus 5 leads.
Reported ±1 standard-error ranges overlap on 13 of 32 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 | Claude Opus 5 | GPT-5.6 Sol | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals Index | 63.67% ±0.96 | 58.01% ±1.02 | 5.67 pts | Reported ±1 SE ranges do not overlap |
| Vals RSI Index | 33.02% | 23.88% | 9.14 pts | Uncertainty comparison unavailable |
| Harvey's Legal Agent Benchmark | 6.67% ±1.65 | 2.50% ±0.83 | 4.17 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 |
| LegalBench | 86.97% ±0.42 | 86.97% ±0.41 | 0.01 pts | Reported ±1 SE ranges overlap |
| EMB | 73.56% ±2.24 | 72.34% ±2.37 | 1.22 pts | Reported ±1 SE ranges overlap |
| Finance Agent (v2) | 58.63% ±0.08 | 53.76% ±0.85 | 4.88 pts | Reported ±1 SE ranges do not overlap |
| MortgageTax | 72.06% ±0.88 | 67.29% ±0.92 | 4.77 pts | Reported ±1 SE ranges do not overlap |
| Tax Agent Bench | 75.06% ±2.96 | 67.95% ±3.12 | 7.11 pts | Reported ±1 SE ranges do not overlap |
| TaxEval v2 | 75.14% ±0.83 | 74.78% ±0.86 | 0.37 pts | Reported ±1 SE ranges overlap |
| MedCode | 63.57% ±1.99 | 43.97% ±2.26 | 19.60 pts | Reported ±1 SE ranges do not overlap |
| MedScribe | 90.98% ±1.92 | 85.23% ±1.97 | 5.75 pts | Reported ±1 SE ranges do not overlap |
| ProofBench v1.1 | 99.00% ±1.00 | 83.00% ±3.77 | 16.00 pts | Reported ±1 SE ranges do not overlap |
| BioMysteryBench | 79.26% ±2.06 | 71.11% ±0.64 | 8.15 pts | Reported ±1 SE ranges do not overlap |
| MysteryMechanism | 37.39% ±3.25 | 33.33% ±3.17 | 4.05 pts | Reported ±1 SE ranges overlap |
| Terminal-Bench Science | 27.14% ±5.35 | 20.00% ±4.82 | 7.14 pts | Reported ±1 SE ranges overlap |
| GPQA Diamond | 93.43% ±1.24 | 95.20% ±1.07 | 1.77 pts | Reported ±1 SE ranges overlap |
| MMLU Pro | 91.59% ±0.28 | 89.10% ±0.31 | 2.49 pts | Reported ±1 SE ranges do not overlap |
| MMMU Pro | 89.88% ±0.72 | 88.84% ±0.76 | 1.04 pts | Reported ±1 SE ranges overlap |
| SAGE | 49.43% ±3.29 | 52.56% ±3.42 | 3.14 pts | Reported ±1 SE ranges overlap |
| Code Migration | 57.47% ±4.37 | 52.92% ±4.35 | 4.56 pts | Reported ±1 SE ranges overlap |
| IOI | 84.33% ±9.96 | 91.17% ±4.51 | 6.83 pts | Reported ±1 SE ranges overlap |
| LiveCodeBench | 89.03% ±0.91 | 82.60% ±1.09 | 6.43 pts | Reported ±1 SE ranges do not overlap |
| ProgramBench | 3.00% ±1.21 | 1.50% ±0.86 | 1.50 pts | Reported ±1 SE ranges overlap |
| SkillsBench | 60.44% ±4.58 | 54.10% ±4.73 | 6.34 pts | Reported ±1 SE ranges overlap |
| SWE-bench | 97.00% ±0.76 | 96.20% ±0.86 | 0.80 pts | Reported ±1 SE ranges overlap |
| Terminal-Bench 4.0 | 53.53% ±1.34 | 37.88% ±0.00 | 15.66 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench 1-100 | 28.53% ±4.33 | 20.01% ±3.63 | 8.52 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 88.40% ±3.00 | 80.50% ±3.72 | 7.91 pts | Reported ±1 SE ranges do not overlap |
| CyberBench v1.1 | 65.36% ±5.55 | 76.31% ±5.18 | 10.95 pts | Reported ±1 SE ranges do not overlap |
| SRE Bench | 12.21% ±2.03 | 30.53% ±2.85 | 18.32 pts | Reported ±1 SE ranges do not overlap |
| Public Benefits Bench v1.1 | 76.93% ±1.10 | 66.51% ±1.23 | 10.42 pts | Reported ±1 SE ranges do not overlap |
| CUA-bench | 9.00% | 8.33% | 0.67 pts | Uncertainty comparison unavailable |
| Time Horizon Index: KSP | 18.83% ±0.00 | 23.83% ±0.00 | 5.00 pts | Reported ±1 SE ranges do not overlap |
Performance by category
| Category | Claude Opus 5 average | GPT-5.6 Sol average |
|---|---|---|
| Legal | 49.64% | 45.85% |
| Finance | 70.89% | 67.22% |
| Healthcare | 77.28% | 64.60% |
| Math | 99.00% | 83.00% |
| Science | 47.93% | 41.48% |
| Academic | 91.64% | 91.05% |
| Education | 49.43% | 52.56% |
| Coding | 62.42% | 57.43% |
| Cyber | 38.79% | 53.42% |
| Social Mobility | 76.93% | 66.51% |
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 | Claude Opus 5 cost | GPT-5.6 Sol cost | Claude Opus 5 latency | GPT-5.6 Sol latency |
|---|---|---|---|---|
| Vals Index | $19.31 | $14.24 | 57m41s | 37m23s |
| Vals RSI Index | $418.69 | $347.10 | 90h00m | 90h00m |
| Harvey's Legal Agent Benchmark | $23.67 | $10.37 | 55m37s | 22m50s |
| Legal Research Bench | $6.76 | $21.61 | 31m01s | 1h17m |
| LegalBench | N/A | N/A | 4.42s | 6.20s |
| EMB | $6.07 | $6.01 | 25m59s | 9m50s |
| Finance Agent (v2) | $5.12 | $1.25 | 9m58s | 19m25s |
| MortgageTax | N/A | N/A | 7.98s | 12.62s |
| Tax Agent Bench | $5.14 | $7.51 | 16m48s | 44m41s |
| TaxEval v2 | N/A | N/A | 58.34s | 62.03s |
| MedCode | N/A | N/A | 24.50s | 96.59s |
| MedScribe | N/A | N/A | 76.56s | 94.40s |
| ProofBench v1.1 | $2.11 | $1.44 | 9m51s | 8m40s |
| BioMysteryBench | $3.29 | $1.93 | 36m36s | 18m18s |
| MysteryMechanism | $3.28 | $0.96 | 16m53s | 8m43s |
| Terminal-Bench Science | $32.54 | $7.32 | 2h12m | 1h44m |
| GPQA Diamond | N/A | N/A | 39.59s | 58.30s |
| MMLU Pro | N/A | N/A | 9.60s | 15.65s |
| MMMU Pro | N/A | N/A | 30.03s | 26.20s |
| SAGE | N/A | N/A | 56.17s | 75.39s |
| Code Migration | $60.51 | $24.54 | 2h51m | 59m12s |
| IOI | $16.48 | $7.98 | 56m42s | 1h02m |
| LiveCodeBench | N/A | N/A | 59.22s | 56.51s |
| ProgramBench | $60.29 | $15.29 | 2h55m | 29m24s |
| SkillsBench | $2.51 | $4.14 | 8m25s | 13m07s |
| SWE-bench | $1.29 | $1.15 | 9m37s | 3m02s |
| Terminal-Bench 4.0 | $18.60 | $7.98 | 1h07m | 33m46s |
| Vibe Code Bench 1-100 | $41.49 | $47.43 | 1h22m | 2h37m |
| Vibe Code Bench v1.1 | $33.88 | $33.40 | 1h28m | 33m53s |
| CyberBench v1.1 | $2.87 | $3.05 | 19m25s | 10m47s |
| SRE Bench | $23.63 | $42.52 | 1h21m | 1h28m |
| Public Benefits Bench v1.1 | $3.95 | $8.85 | 30m09s | 1h01m |
| CUA-bench | $232.23 | $206.99 | 25.29s | 18.80s |
| Time Horizon Index: KSP | $1348.77 | $2027.20 | N/A | N/A |