Gemini 4 Argon vs Wonder Full: Benchmark Comparison
Gemini 4 Argon has the higher score on 21 of 28 shared benchmarks; Wonder Full leads on 7.
The largest observed score gap is 32.44 pts on SRE Bench , where Gemini 4 Argon leads.
Reported ±1 standard-error ranges overlap on 11 of 26 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 | Gemini 4 Argon | Wonder Full | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals Finance Index | 64.89% ±1.49 | 64.55% ±1.40 | 0.34 pts | Reported ±1 SE ranges overlap |
| Vals Index | 69.24% ±1.09 | 66.79% ±1.07 | 2.45 pts | Reported ±1 SE ranges do not overlap |
| Vals Legal Index | 76.72% ±1.84 | 77.86% ±1.82 | 1.15 pts | Reported ±1 SE ranges overlap |
| Vals RSI Index | 30.55% | 24.43% | 6.12 pts | Uncertainty comparison unavailable |
| Vals Science Index | 55.36% ±2.29 | 48.54% ±2.23 | 6.82 pts | Reported ±1 SE ranges do not overlap |
| HLAB-Vals-1.1 | 27.08% ±3.74 | 23.33% ±3.55 | 3.75 pts | Reported ±1 SE ranges overlap |
| Legal Research Bench | 54.81% ±3.46 | 53.37% ±3.47 | 1.44 pts | Reported ±1 SE ranges overlap |
| LegalBench | 88.30% ±0.37 | 87.42% ±0.41 | 0.88 pts | Reported ±1 SE ranges do not overlap |
| EMB v1.1 | 74.83% ±2.49 | 75.00% ±2.14 | 0.17 pts | Reported ±1 SE ranges overlap |
| Finance Agent v2.1 | 67.66% ±2.00 | 67.58% ±1.98 | 0.08 pts | Reported ±1 SE ranges overlap |
| Tax Agent Bench | 44.56% ±3.59 | 43.00% ±3.57 | 1.56 pts | Reported ±1 SE ranges overlap |
| MedCode | 58.80% ±2.10 | 49.93% ±2.18 | 8.87 pts | Reported ±1 SE ranges do not overlap |
| MedScribe | 87.43% ±1.96 | 89.33% ±2.00 | 1.91 pts | Reported ±1 SE ranges overlap |
| ProofBench v1.1 | 99.00% ±1.00 | 94.00% ±2.39 | 5.00 pts | Reported ±1 SE ranges do not overlap |
| BioMysteryBench | 76.30% ±0.37 | 73.33% ±0.64 | 2.96 pts | Reported ±1 SE ranges do not overlap |
| MysteryMechanism | 45.49% ±3.35 | 35.13% ±3.21 | 10.36 pts | Reported ±1 SE ranges do not overlap |
| Robo-bench | 0.83% | 10.42% | 9.58 pts | Uncertainty comparison unavailable |
| Terminal-Bench-Science | 44.29% ±5.98 | 37.14% ±5.82 | 7.14 pts | Reported ±1 SE ranges overlap |
| Code Migration v1.1 | 70.80% ±3.72 | 62.79% ±3.17 | 8.01 pts | Reported ±1 SE ranges do not overlap |
| IOI | 100.00% ±0.00 | 79.83% ±2.99 | 20.17 pts | Reported ±1 SE ranges do not overlap |
| ProgramBench | 2.50% ±1.11 | 0.50% ±0.50 | 2.00 pts | Reported ±1 SE ranges do not overlap |
| SWE-bench Multimodal | 56.04% ±2.27 | 47.29% ±2.28 | 8.75 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench 4.0 | 57.58% ±2.31 | 58.08% ±3.31 | 0.51 pts | Reported ±1 SE ranges overlap |
| Vibe Code Bench 1-100 | 21.20% ±3.81 | 33.98% ±5.32 | 12.78 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 91.91% ±1.90 | 86.00% ±1.93 | 5.91 pts | Reported ±1 SE ranges do not overlap |
| CyberBench v1.1 | 77.86% ±5.30 | 73.63% ±5.56 | 4.23 pts | Reported ±1 SE ranges overlap |
| SRE Bench | 44.27% ±3.08 | 11.83% ±2.00 | 32.44 pts | Reported ±1 SE ranges do not overlap |
| Public Benefits Bench v1.1 | 69.76% ±0.00 | 73.75% ±1.15 | 3.99 pts | Reported ±1 SE ranges do not overlap |
Performance by category
| Category | Gemini 4 Argon average | Wonder Full average |
|---|---|---|
| Legal | 56.73% | 54.71% |
| Finance | 62.35% | 61.86% |
| Healthcare | 73.11% | 69.63% |
| Math | 99.00% | 94.00% |
| Science | 41.73% | 39.01% |
| Coding | 57.15% | 52.64% |
| Cyber | 61.07% | 42.73% |
| Social Mobility | 69.76% | 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 | Gemini 4 Argon cost | Wonder Full cost | Gemini 4 Argon latency | Wonder Full latency |
|---|---|---|---|---|
| Vals Finance Index | $6.53 | $3.14 | 22m41s | 20m06s |
| Vals Index | $13.37 | $6.81 | 44m58s | 59m19s |
| Vals Legal Index | $6.82 | $3.72 | 25m41s | 29m56s |
| Vals RSI Index | N/A | N/A | 90h00m | 90h00m |
| Vals Science Index | $18.40 | $10.15 | 57m00s | 1h34m |
| HLAB-Vals-1.1 | $10.55 | $5.41 | 29m15s | 44m09s |
| Legal Research Bench | $6.45 | $3.60 | 31m21s | 27m34s |
| LegalBench | N/A | N/A | 12.62s | 38.27s |
| EMB v1.1 | $11.80 | $4.58 | 38m26s | 28m05s |
| Finance Agent v2.1 | $4.32 | $2.68 | 13m13s | 14m10s |
| Tax Agent Bench | $3.47 | $2.15 | 16m26s | 18m04s |
| MedCode | N/A | N/A | 79.53s | 3m03s |
| MedScribe | N/A | N/A | 87.44s | 115.72s |
| ProofBench v1.1 | $0.78 | $0.86 | 6m20s | 11m36s |
| BioMysteryBench | $5.71 | $3.47 | 39m57s | 26m14s |
| MysteryMechanism | $4.59 | $4.16 | 24m55s | 41m34s |
| Robo-bench | $17.33 | $19.68 | 2m32s | 2m14s |
| Terminal-Bench-Science | $44.90 | $22.81 | 1h46m | 3h35m |
| Code Migration v1.1 | $44.79 | $23.58 | 2h22m | 4h18m |
| IOI | $5.31 | $7.08 | 24m19s | 1h31m |
| ProgramBench | $17.66 | $28.65 | 1h11m | 6h04m |
| SWE-bench Multimodal | $11.40 | $1.86 | 33m38s | 14m22s |
| Terminal-Bench 4.0 | $17.64 | $10.53 | 53m35s | 1h12m |
| Vibe Code Bench 1-100 | $18.66 | $19.54 | 1h43m | 1h57m |
| Vibe Code Bench v1.1 | $8.11 | $2.29 | 37m28s | 10m48s |
| CyberBench v1.1 | $5.11 | $3.63 | 19m10s | 35m27s |
| SRE Bench | $30.45 | $33.81 | 1h35m | 5h34m |
| Public Benefits Bench v1.1 | $2.79 | $4.64 | 26m07s | 1h13m |
Results available only for Gemini 4 Argon
- SAGE
- CUA-bench
- Time Horizon Index: KSP
Results available only for Wonder Full
None.