Wonder Full vs GPT-6.1 Sol: Benchmark Comparison
Wonder Full has the higher score on 14 of 26 shared benchmarks; GPT-6.1 Sol leads on 12.
The largest observed score gap is 38.93 pts on SRE Bench , where GPT-6.1 Sol leads.
Reported ±1 standard-error ranges overlap on 6 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 | Wonder Full | GPT-6.1 Sol | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals Finance Index | 64.55% ±1.40 | 55.81% ±1.50 | 8.74 pts | Reported ±1 SE ranges do not overlap |
| Vals Index | 66.79% ±1.07 | 63.94% ±1.07 | 2.84 pts | Reported ±1 SE ranges do not overlap |
| Vals Legal Index | 77.86% ±1.82 | 68.72% ±1.77 | 9.14 pts | Reported ±1 SE ranges do not overlap |
| Vals RSI Index | 24.43% | 26.62% | 2.19 pts | Uncertainty comparison unavailable |
| Vals Science Index | 48.54% ±2.23 | 59.63% ±2.31 | 11.09 pts | Reported ±1 SE ranges do not overlap |
| HLAB-Vals-1.1 | 23.33% ±3.55 | 6.25% ±1.83 | 17.08 pts | Reported ±1 SE ranges do not overlap |
| Legal Research Bench | 53.37% ±3.47 | 38.46% ±3.38 | 14.90 pts | Reported ±1 SE ranges do not overlap |
| EMB v1.1 | 75.00% ±2.14 | 71.09% ±2.75 | 3.91 pts | Reported ±1 SE ranges overlap |
| Finance Agent v2.1 | 67.58% ±1.98 | 61.82% ±2.11 | 5.76 pts | Reported ±1 SE ranges do not overlap |
| Tax Agent Bench | 43.00% ±3.57 | 21.75% ±2.98 | 21.24 pts | Reported ±1 SE ranges do not overlap |
| MedCode | 49.93% ±2.18 | 48.84% ±2.12 | 1.09 pts | Reported ±1 SE ranges overlap |
| MedScribe | 89.33% ±2.00 | 86.45% ±1.90 | 2.88 pts | Reported ±1 SE ranges overlap |
| ProofBench v1.1 | 94.00% ±2.39 | 99.00% ±1.00 | 5.00 pts | Reported ±1 SE ranges do not overlap |
| BioMysteryBench | 73.33% ±0.64 | 79.63% ±0.74 | 6.30 pts | Reported ±1 SE ranges do not overlap |
| MysteryMechanism | 35.13% ±3.21 | 46.40% ±3.35 | 11.26 pts | Reported ±1 SE ranges do not overlap |
| Robo-bench | 10.42% | 28.33% | 17.92 pts | Uncertainty comparison unavailable |
| Terminal-Bench-Science | 37.14% ±5.82 | 52.86% ±6.01 | 15.71 pts | Reported ±1 SE ranges do not overlap |
| Code Migration v1.1 | 62.79% ±3.17 | 77.32% ±2.98 | 14.53 pts | Reported ±1 SE ranges do not overlap |
| IOI | 79.83% ±2.99 | 96.89% ±3.11 | 17.06 pts | Reported ±1 SE ranges do not overlap |
| ProgramBench | 0.50% ±0.50 | 3.00% ±1.21 | 2.50 pts | Reported ±1 SE ranges do not overlap |
| SWE-bench Multimodal | 47.29% ±2.28 | 43.54% ±2.27 | 3.75 pts | Reported ±1 SE ranges overlap |
| Terminal-Bench 4.0 | 58.08% ±3.31 | 55.05% ±1.82 | 3.03 pts | Reported ±1 SE ranges overlap |
| Vibe Code Bench v1.1 | 86.00% ±1.93 | 88.93% ±1.89 | 2.94 pts | Reported ±1 SE ranges overlap |
| CyberBench v1.1 | 73.63% ±5.56 | 39.29% ±2.74 | 34.34 pts | Reported ±1 SE ranges do not overlap |
| SRE Bench | 11.83% ±2.00 | 50.76% ±3.10 | 38.93 pts | Reported ±1 SE ranges do not overlap |
| Public Benefits Bench v1.1 | 73.75% ±1.15 | 59.34% ±1.28 | 14.41 pts | Reported ±1 SE ranges do not overlap |
Performance by category
| Category | Wonder Full average | GPT-6.1 Sol average |
|---|---|---|
| Legal | 38.35% | 22.36% |
| Finance | 61.86% | 51.55% |
| Healthcare | 69.63% | 67.65% |
| Math | 94.00% | 99.00% |
| Science | 39.01% | 51.80% |
| Coding | 55.75% | 60.79% |
| Cyber | 42.73% | 45.02% |
| Social Mobility | 73.75% | 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 Full cost | GPT-6.1 Sol cost | Wonder Full latency | GPT-6.1 Sol latency |
|---|---|---|---|---|
| Vals Finance Index | $3.14 | $1.36 | 20m06s | 19m16s |
| Vals Index | $6.81 | $3.61 | 59m19s | 44m24s |
| Vals Legal Index | $3.72 | $2.75 | 29m56s | 31m40s |
| Vals RSI Index | N/A | N/A | 90h00m | 90h00m |
| Vals Science Index | $10.15 | $1.36 | 1h34m | 33m11s |
| HLAB-Vals-1.1 | $5.41 | $3.54 | 44m09s | 40m26s |
| Legal Research Bench | $3.60 | $3.38 | 27m34s | 34m20s |
| EMB v1.1 | $4.58 | $1.14 | 28m05s | 18m28s |
| Finance Agent v2.1 | $2.68 | $1.63 | 14m10s | 19m05s |
| Tax Agent Bench | $2.15 | $1.32 | 18m04s | 20m14s |
| MedCode | N/A | N/A | 3m03s | 115.34s |
| MedScribe | N/A | N/A | 115.72s | 2m50s |
| ProofBench v1.1 | $0.86 | $0.23 | 11m36s | 4m45s |
| BioMysteryBench | $3.47 | $0.37 | 26m14s | 8m51s |
| MysteryMechanism | $4.16 | $0.28 | 41m34s | 7m17s |
| Robo-bench | $19.68 | $4.97 | 2m14s | 79.89s |
| Terminal-Bench-Science | $22.81 | $3.44 | 3h35m | 1h23m |
| Code Migration v1.1 | $23.58 | $9.51 | 4h18m | 1h39m |
| IOI | $7.08 | $1.26 | 1h31m | 31m31s |
| ProgramBench | $28.65 | $2.87 | 6h04m | 1h10m |
| SWE-bench Multimodal | $1.86 | $0.35 | 14m22s | 6m56s |
| Terminal-Bench 4.0 | $10.53 | $1.72 | 1h12m | 45m12s |
| Vibe Code Bench v1.1 | $2.29 | $6.23 | 10m48s | 1h15m |
| CyberBench v1.1 | $3.63 | $0.29 | 35m27s | 3m42s |
| SRE Bench | $33.81 | $2.69 | 5h34m | 32m32s |
| Public Benefits Bench v1.1 | $4.64 | $0.69 | 1h13m | 50m24s |
Results available only for Wonder Full
- LegalBench
- Vibe Code Bench 1-100
Results available only for GPT-6.1 Sol
- ClaimBench
- SAGE