Claude Sonnet 5.5 vs Wonder Full: Benchmark Comparison
Claude Sonnet 5.5 has the higher score on 16 of 24 shared benchmarks; Wonder Full leads on 8.
The largest observed score gap is 24.09 pts on Code Migration v1.1 , where Claude Sonnet 5.5 leads.
Reported ±1 standard-error ranges overlap on 9 of 24 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 Full | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals Finance Index | 64.53% ±1.48 | 64.55% ±1.40 | 0.02 pts | Reported ±1 SE ranges overlap |
| Vals Index | 70.85% ±0.93 | 66.79% ±1.07 | 4.07 pts | Reported ±1 SE ranges do not overlap |
| Vals Legal Index | 68.02% ±1.81 | 77.86% ±1.82 | 9.84 pts | Reported ±1 SE ranges do not overlap |
| Vals Science Index | 59.12% ±2.31 | 48.54% ±2.23 | 10.58 pts | Reported ±1 SE ranges do not overlap |
| HLAB-Vals-1.1 | 10.42% ±2.57 | 23.33% ±3.55 | 12.92 pts | Reported ±1 SE ranges do not overlap |
| Legal Research Bench | 48.08% ±3.47 | 53.37% ±3.47 | 5.29 pts | Reported ±1 SE ranges overlap |
| EMB v1.1 | 75.90% ±2.47 | 75.00% ±2.14 | 0.90 pts | Reported ±1 SE ranges overlap |
| Finance Agent v2.1 | 67.28% ±2.02 | 67.58% ±1.98 | 0.30 pts | Reported ±1 SE ranges overlap |
| Tax Agent Bench | 41.96% ±3.56 | 43.00% ±3.57 | 1.04 pts | Reported ±1 SE ranges overlap |
| MedCode | 52.92% ±2.12 | 49.93% ±2.18 | 2.99 pts | Reported ±1 SE ranges overlap |
| MedScribe | 91.10% ±1.96 | 89.33% ±2.00 | 1.77 pts | Reported ±1 SE ranges overlap |
| ProofBench v1.1 | 100.00% ±0.00 | 94.00% ±2.39 | 6.00 pts | Reported ±1 SE ranges do not overlap |
| BioMysteryBench | 81.11% ±0.64 | 73.33% ±0.64 | 7.78 pts | Reported ±1 SE ranges do not overlap |
| MysteryMechanism | 49.10% ±3.36 | 35.13% ±3.21 | 13.96 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench-Science | 47.14% ±6.01 | 37.14% ±5.82 | 10.00 pts | Reported ±1 SE ranges overlap |
| Code Migration v1.1 | 86.88% ±1.52 | 62.79% ±3.17 | 24.09 pts | Reported ±1 SE ranges do not overlap |
| IOI | 83.06% ±3.74 | 79.83% ±2.99 | 3.22 pts | Reported ±1 SE ranges overlap |
| ProgramBench | 6.50% ±1.75 | 0.50% ±0.50 | 6.00 pts | Reported ±1 SE ranges do not overlap |
| SWE-bench Multimodal | 58.33% ±2.25 | 47.29% ±2.28 | 11.04 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench 4.0 | 64.14% ±1.01 | 58.08% ±3.31 | 6.06 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 92.39% ±1.26 | 86.00% ±1.93 | 6.40 pts | Reported ±1 SE ranges do not overlap |
| CyberBench v1.1 | 59.58% ±5.21 | 73.63% ±5.56 | 14.05 pts | Reported ±1 SE ranges do not overlap |
| SRE Bench | 30.15% ±2.84 | 11.83% ±2.00 | 18.32 pts | Reported ±1 SE ranges do not overlap |
| Public Benefits Bench v1.1 | 67.19% ±1.22 | 73.75% ±1.15 | 6.56 pts | Reported ±1 SE ranges do not overlap |
Performance by category
| Category | Claude Sonnet 5.5 average | Wonder Full average |
|---|---|---|
| Legal | 29.25% | 38.35% |
| Finance | 61.71% | 61.86% |
| Healthcare | 72.01% | 69.63% |
| Math | 100.00% | 94.00% |
| Science | 59.12% | 48.54% |
| Coding | 65.22% | 55.75% |
| Cyber | 44.87% | 42.73% |
| Social Mobility | 67.19% | 73.75% |
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 Full cost | Claude Sonnet 5.5 latency | Wonder Full latency |
|---|---|---|---|---|
| Vals Finance Index | $8.33 | $3.14 | 43m31s | 20m06s |
| Vals Index | $18.70 | $6.81 | 1h26m | 59m19s |
| Vals Legal Index | $13.56 | $3.72 | 1h15m | 29m56s |
| Vals Science Index | $13.24 | $10.15 | 1h09m | 1h34m |
| HLAB-Vals-1.1 | $16.61 | $5.41 | 1h13m | 44m09s |
| Legal Research Bench | $14.83 | $3.60 | 1h33m | 27m34s |
| EMB v1.1 | $8.79 | $4.58 | 36m45s | 28m05s |
| Finance Agent v2.1 | $6.95 | $2.68 | 34m26s | 14m10s |
| Tax Agent Bench | $9.25 | $2.15 | 59m21s | 18m04s |
| MedCode | N/A | N/A | 3m59s | 3m03s |
| MedScribe | N/A | N/A | 5m07s | 115.72s |
| ProofBench v1.1 | $0.56 | $0.86 | 5m48s | 11m36s |
| BioMysteryBench | $3.74 | $3.47 | 18m06s | 26m14s |
| MysteryMechanism | $3.45 | $4.16 | 23m30s | 41m34s |
| Terminal-Bench-Science | $32.52 | $22.81 | 2h46m | 3h35m |
| Code Migration v1.1 | $43.88 | $23.58 | 4h07m | 4h18m |
| IOI | $7.52 | $7.08 | 45m51s | 1h31m |
| ProgramBench | $35.86 | $28.65 | 2h07m | 6h04m |
| SWE-bench Multimodal | $4.53 | $1.86 | 24m13s | 14m22s |
| Terminal-Bench 4.0 | $16.51 | $10.53 | 1h22m | 1h12m |
| Vibe Code Bench v1.1 | $31.25 | $2.29 | 1h01m | 10m48s |
| CyberBench v1.1 | $4.42 | $3.63 | 33m53s | 35m27s |
| SRE Bench | $26.52 | $33.81 | 1h55m | 5h34m |
| Public Benefits Bench v1.1 | $4.25 | $4.64 | 1h15m | 1h13m |
Results available only for Claude Sonnet 5.5
- ClaimBench
- SAGE
Results available only for Wonder Full
- Vals RSI Index
- LegalBench
- Robo-bench
- Vibe Code Bench 1-100