Claude Opus 5 vs Wonder: Benchmark Comparison
Claude Opus 5 has the higher score on 13 of 14 shared benchmarks; Wonder leads on 1.
The largest observed score gap is 17.39 pts on IOI , where Claude Opus 5 leads.
Reported ±1 standard-error ranges overlap on 4 of 13 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 | Wonder | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals Science Index | 47.93% ±2.20 | 39.12% ±1.96 | 8.81 pts | Reported ±1 SE ranges do not overlap |
| Legal Research Bench | 55.29% ±3.46 | 50.48% ±3.48 | 4.81 pts | Reported ±1 SE ranges overlap |
| Tax Agent Bench | 45.55% ±3.59 | 36.27% ±3.47 | 9.28 pts | Reported ±1 SE ranges do not overlap |
| ProofBench v1.1 | 99.00% ±1.00 | 84.00% ±3.69 | 15.00 pts | Reported ±1 SE ranges do not overlap |
| BioMysteryBench | 79.26% ±2.06 | 67.04% ±0.37 | 12.22 pts | Reported ±1 SE ranges do not overlap |
| MysteryMechanism | 37.39% ±3.25 | 27.48% ±3.00 | 9.91 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench-Science | 27.14% ±5.35 | 22.86% ±5.05 | 4.29 pts | Reported ±1 SE ranges overlap |
| IOI | 84.33% ±9.96 | 66.94% ±4.91 | 17.39 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench 4.0 | 53.53% ±1.34 | 49.49% ±2.20 | 4.04 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench 1-100 | 28.53% ±4.33 | 26.55% ±4.33 | 1.98 pts | Reported ±1 SE ranges overlap |
| Vibe Code Bench v1.1 | 88.40% ±3.00 | 83.09% ±2.49 | 5.31 pts | Reported ±1 SE ranges overlap |
| CyberBench v1.1 | 65.36% ±5.55 | 80.36% ±5.13 | 15.00 pts | Reported ±1 SE ranges do not overlap |
| Public Benefits Bench v1.1 | 76.93% ±1.10 | 70.23% ±1.19 | 6.70 pts | Reported ±1 SE ranges do not overlap |
| CUA-bench | 9.00% | 7.17% | 1.83 pts | Uncertainty comparison unavailable |
Performance by category
| Category | Claude Opus 5 average | Wonder average |
|---|---|---|
| Legal | 55.29% | 50.48% |
| Finance | 45.55% | 36.27% |
| Math | 99.00% | 84.00% |
| Science | 47.93% | 39.12% |
| Coding | 63.70% | 56.52% |
| Cyber | 65.36% | 80.36% |
| Social Mobility | 76.93% | 70.23% |
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 | Wonder cost | Claude Opus 5 latency | Wonder latency |
|---|---|---|---|---|
| Vals Science Index | $13.04 | $8.74 | 1h02m | 43m08s |
| Legal Research Bench | $6.76 | $3.98 | 31m01s | 15m27s |
| Tax Agent Bench | $5.14 | $1.88 | 16m48s | 8m13s |
| ProofBench v1.1 | $2.11 | $1.31 | 9m51s | 15m29s |
| BioMysteryBench | $3.29 | $3.23 | 36m36s | 18m02s |
| MysteryMechanism | $3.28 | $3.39 | 16m53s | 18m17s |
| Terminal-Bench-Science | $32.54 | $19.61 | 2h12m | 1h33m |
| IOI | $16.48 | $4.54 | 56m42s | 25m02s |
| Terminal-Bench 4.0 | $18.60 | $14.66 | 1h07m | 54m41s |
| Vibe Code Bench 1-100 | $41.49 | $13.72 | 1h22m | 31m08s |
| Vibe Code Bench v1.1 | $33.88 | $1.80 | 1h28m | 5m50s |
| CyberBench v1.1 | $2.87 | $3.66 | 19m25s | 13m45s |
| Public Benefits Bench v1.1 | $3.95 | $2.46 | 30m09s | 19m44s |
| CUA-bench | $232.23 | $597.21 | 25.29s | 30.56s |
Results available only for Claude Opus 5
- Vals RSI Index
- LegalBench
- MortgageTax
- TaxEval v2
- MedCode
- MedScribe
- GPQA Diamond
- MMLU Pro
- MMMU Pro
- SAGE
- LiveCodeBench
- ProgramBench
- SkillsBench
- SWE-bench
- SRE Bench
- Time Horizon Index: KSP
Results available only for Wonder
- Vals Finance Index
- Vals Index
- Vals Legal Index
- HLAB-Vals-1.1
- EMB v1.1
- Finance Agent v2.1
- Code Migration v1.1
- SWE-bench Multimodal