Wonder vs GPT-6.1 Sol: Benchmark Comparison
Wonder has the higher score on 11 of 20 shared benchmarks; GPT-6.1 Sol leads on 9.
The largest observed score gap is 41.07 pts on CyberBench v1.1 , where Wonder leads.
Reported ±1 standard-error ranges overlap on 5 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 | Wonder | GPT-6.1 Sol | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals Finance Index | 61.23% ±1.46 | 55.81% ±1.50 | 5.41 pts | Reported ±1 SE ranges do not overlap |
| Vals Index | 64.88% ±1.07 | 63.94% ±1.07 | 0.93 pts | Reported ±1 SE ranges overlap |
| Vals Legal Index | 75.26% ±1.83 | 68.72% ±1.77 | 6.54 pts | Reported ±1 SE ranges do not overlap |
| Vals Science Index | 39.12% ±1.96 | 59.63% ±2.31 | 20.50 pts | Reported ±1 SE ranges do not overlap |
| HLAB-Vals-1.1 | 23.75% ±3.65 | 6.25% ±1.83 | 17.50 pts | Reported ±1 SE ranges do not overlap |
| Legal Research Bench | 50.48% ±3.48 | 38.46% ±3.38 | 12.02 pts | Reported ±1 SE ranges do not overlap |
| EMB v1.1 | 72.50% ±2.41 | 71.09% ±2.75 | 1.42 pts | Reported ±1 SE ranges overlap |
| Finance Agent v2.1 | 65.55% ±2.03 | 61.82% ±2.11 | 3.73 pts | Reported ±1 SE ranges overlap |
| Tax Agent Bench | 36.27% ±3.47 | 21.75% ±2.98 | 14.51 pts | Reported ±1 SE ranges do not overlap |
| ProofBench v1.1 | 84.00% ±3.69 | 99.00% ±1.00 | 15.00 pts | Reported ±1 SE ranges do not overlap |
| BioMysteryBench | 67.04% ±0.37 | 79.63% ±0.74 | 12.59 pts | Reported ±1 SE ranges do not overlap |
| MysteryMechanism | 27.48% ±3.00 | 46.40% ±3.35 | 18.92 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench-Science | 22.86% ±5.05 | 52.86% ±6.01 | 30.00 pts | Reported ±1 SE ranges do not overlap |
| Code Migration v1.1 | 72.36% ±3.19 | 77.32% ±2.98 | 4.97 pts | Reported ±1 SE ranges overlap |
| IOI | 66.94% ±4.91 | 96.89% ±3.11 | 29.94 pts | Reported ±1 SE ranges do not overlap |
| SWE-bench Multimodal | 45.63% ±2.28 | 43.54% ±2.27 | 2.08 pts | Reported ±1 SE ranges overlap |
| Terminal-Bench 4.0 | 49.49% ±2.20 | 55.05% ±1.82 | 5.56 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 83.09% ±2.49 | 88.93% ±1.89 | 5.84 pts | Reported ±1 SE ranges do not overlap |
| CyberBench v1.1 | 80.36% ±5.13 | 39.29% ±2.74 | 41.07 pts | Reported ±1 SE ranges do not overlap |
| Public Benefits Bench v1.1 | 70.23% ±1.19 | 59.34% ±1.28 | 10.89 pts | Reported ±1 SE ranges do not overlap |
Performance by category
| Category | Wonder average | GPT-6.1 Sol average |
|---|---|---|
| Legal | 37.12% | 22.36% |
| Finance | 58.11% | 51.55% |
| Math | 84.00% | 99.00% |
| Science | 39.12% | 59.63% |
| Coding | 63.50% | 72.35% |
| Cyber | 80.36% | 39.29% |
| Social Mobility | 70.23% | 59.34% |
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 | Wonder cost | GPT-6.1 Sol cost | Wonder latency | GPT-6.1 Sol latency |
|---|---|---|---|---|
| Vals Finance Index | $3.37 | $1.36 | 10m12s | 19m16s |
| Vals Index | $10.65 | $3.61 | 35m24s | 44m24s |
| Vals Legal Index | $4.32 | $2.75 | 12m55s | 31m40s |
| Vals Science Index | $8.74 | $1.36 | 43m08s | 33m11s |
| HLAB-Vals-1.1 | $7.08 | $3.54 | 15m06s | 40m26s |
| Legal Research Bench | $3.98 | $3.38 | 15m27s | 34m20s |
| EMB v1.1 | $5.60 | $1.14 | 17m14s | 18m28s |
| Finance Agent v2.1 | $2.62 | $1.63 | 5m10s | 19m05s |
| Tax Agent Bench | $1.88 | $1.32 | 8m13s | 20m14s |
| ProofBench v1.1 | $1.31 | $0.23 | 15m29s | 4m45s |
| BioMysteryBench | $3.23 | $0.37 | 18m02s | 8m51s |
| MysteryMechanism | $3.39 | $0.28 | 18m17s | 7m17s |
| Terminal-Bench-Science | $19.61 | $3.44 | 1h33m | 1h23m |
| Code Migration v1.1 | $47.93 | $9.51 | 2h42m | 1h39m |
| IOI | $4.54 | $1.26 | 25m02s | 31m31s |
| SWE-bench Multimodal | $1.70 | $0.35 | 9m44s | 6m56s |
| Terminal-Bench 4.0 | $14.66 | $1.72 | 54m41s | 45m12s |
| Vibe Code Bench v1.1 | $1.80 | $6.23 | 5m50s | 1h15m |
| CyberBench v1.1 | $3.66 | $0.29 | 13m45s | 3m42s |
| Public Benefits Bench v1.1 | $2.46 | $0.69 | 19m44s | 50m24s |
Results available only for Wonder
- Vibe Code Bench 1-100
- CUA-bench
Results available only for GPT-6.1 Sol
- Vals RSI Index
- ClaimBench
- MedCode
- MedScribe
- Robo-bench
- SAGE
- ProgramBench
- SRE Bench