Claude Sonnet 5.5 vs Wonder: Benchmark Comparison
Claude Sonnet 5.5 has the higher score on 15 of 20 shared benchmarks; Wonder leads on 5.
The largest observed score gap is 24.29 pts on Terminal-Bench-Science , where Claude Sonnet 5.5 leads.
Reported ±1 standard-error ranges overlap on 4 of 20 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 Sonnet 5.5 | Wonder | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals Finance Index | 64.53% ±1.48 | 61.23% ±1.46 | 3.31 pts | Reported ±1 SE ranges do not overlap |
| Vals Index | 70.85% ±0.93 | 64.88% ±1.07 | 5.98 pts | Reported ±1 SE ranges do not overlap |
| Vals Legal Index | 68.02% ±1.81 | 75.26% ±1.83 | 7.24 pts | Reported ±1 SE ranges do not overlap |
| Vals Science Index | 59.12% ±2.31 | 39.12% ±1.96 | 19.99 pts | Reported ±1 SE ranges do not overlap |
| HLAB-Vals-1.1 | 10.42% ±2.57 | 23.75% ±3.65 | 13.33 pts | Reported ±1 SE ranges do not overlap |
| Legal Research Bench | 48.08% ±3.47 | 50.48% ±3.48 | 2.40 pts | Reported ±1 SE ranges overlap |
| EMB v1.1 | 75.90% ±2.47 | 72.50% ±2.41 | 3.40 pts | Reported ±1 SE ranges overlap |
| Finance Agent v2.1 | 67.28% ±2.02 | 65.55% ±2.03 | 1.73 pts | Reported ±1 SE ranges overlap |
| Tax Agent Bench | 41.96% ±3.56 | 36.27% ±3.47 | 5.69 pts | Reported ±1 SE ranges overlap |
| ProofBench v1.1 | 100.00% ±0.00 | 84.00% ±3.69 | 16.00 pts | Reported ±1 SE ranges do not overlap |
| BioMysteryBench | 81.11% ±0.64 | 67.04% ±0.37 | 14.07 pts | Reported ±1 SE ranges do not overlap |
| MysteryMechanism | 49.10% ±3.36 | 27.48% ±3.00 | 21.62 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench-Science | 47.14% ±6.01 | 22.86% ±5.05 | 24.29 pts | Reported ±1 SE ranges do not overlap |
| Code Migration v1.1 | 86.88% ±1.52 | 72.36% ±3.19 | 14.52 pts | Reported ±1 SE ranges do not overlap |
| IOI | 83.06% ±3.74 | 66.94% ±4.91 | 16.11 pts | Reported ±1 SE ranges do not overlap |
| SWE-bench Multimodal | 58.33% ±2.25 | 45.63% ±2.28 | 12.71 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench 4.0 | 64.14% ±1.01 | 49.49% ±2.20 | 14.65 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 92.39% ±1.26 | 83.09% ±2.49 | 9.30 pts | Reported ±1 SE ranges do not overlap |
| CyberBench v1.1 | 59.58% ±5.21 | 80.36% ±5.13 | 20.77 pts | Reported ±1 SE ranges do not overlap |
| Public Benefits Bench v1.1 | 67.19% ±1.22 | 70.23% ±1.19 | 3.05 pts | Reported ±1 SE ranges do not overlap |
Performance by category
| Category | Claude Sonnet 5.5 average | Wonder average |
|---|---|---|
| Legal | 29.25% | 37.12% |
| Finance | 61.71% | 58.11% |
| Math | 100.00% | 84.00% |
| Science | 59.12% | 39.12% |
| Coding | 76.96% | 63.50% |
| Cyber | 59.58% | 80.36% |
| Social Mobility | 67.19% | 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 Sonnet 5.5 cost | Wonder cost | Claude Sonnet 5.5 latency | Wonder latency |
|---|---|---|---|---|
| Vals Finance Index | $8.33 | $3.37 | 43m31s | 10m12s |
| Vals Index | $18.70 | $10.65 | 1h26m | 35m24s |
| Vals Legal Index | $13.56 | $4.32 | 1h15m | 12m55s |
| Vals Science Index | $13.24 | $8.74 | 1h09m | 43m08s |
| HLAB-Vals-1.1 | $16.61 | $7.08 | 1h13m | 15m06s |
| Legal Research Bench | $14.83 | $3.98 | 1h33m | 15m27s |
| EMB v1.1 | $8.79 | $5.60 | 36m45s | 17m14s |
| Finance Agent v2.1 | $6.95 | $2.62 | 34m26s | 5m10s |
| Tax Agent Bench | $9.25 | $1.88 | 59m21s | 8m13s |
| ProofBench v1.1 | $0.56 | $1.31 | 5m48s | 15m29s |
| BioMysteryBench | $3.74 | $3.23 | 18m06s | 18m02s |
| MysteryMechanism | $3.45 | $3.39 | 23m30s | 18m17s |
| Terminal-Bench-Science | $32.52 | $19.61 | 2h46m | 1h33m |
| Code Migration v1.1 | $43.88 | $47.93 | 4h07m | 2h42m |
| IOI | $7.52 | $4.54 | 45m51s | 25m02s |
| SWE-bench Multimodal | $4.53 | $1.70 | 24m13s | 9m44s |
| Terminal-Bench 4.0 | $16.51 | $14.66 | 1h22m | 54m41s |
| Vibe Code Bench v1.1 | $31.25 | $1.80 | 1h01m | 5m50s |
| CyberBench v1.1 | $4.42 | $3.66 | 33m53s | 13m45s |
| Public Benefits Bench v1.1 | $4.25 | $2.46 | 1h15m | 19m44s |
Results available only for Claude Sonnet 5.5
- ClaimBench
- MedCode
- MedScribe
- SAGE
- ProgramBench
- SRE Bench
Results available only for Wonder
- Vibe Code Bench 1-100
- CUA-bench