Model comparison
Gemini 3.7 Flash vs Grok 4.7: Benchmark Comparison
Gemini 3.7 Flash has the higher score on 7 of 19 shared benchmarks; Grok 4.7 leads on 12.
The largest observed score gap is 32.00 pts on ProofBench v1.1 , where Gemini 3.7 Flash leads.
Reported ±1 standard-error ranges overlap on 6 of 18 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 3.7 Flash | Grok 4.7 | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals Index | 59.31% ±1.06 | 60.22% ±1.08 | 0.91 pts | Reported ±1 SE ranges overlap |
| Vals RSI Index | 18.27% | 20.20% | 1.93 pts | Uncertainty comparison unavailable |
| Harvey's Legal Agent Benchmark | 8.75% ±2.15 | 12.50% ±2.42 | 3.75 pts | Reported ±1 SE ranges overlap |
| Legal Research Bench | 34.62% ±3.31 | 47.12% ±3.47 | 12.50 pts | Reported ±1 SE ranges do not overlap |
| LegalBench | 87.26% ±0.42 | 84.39% ±0.46 | 2.87 pts | Reported ±1 SE ranges do not overlap |
| EMB | 71.33% ±2.26 | 66.99% ±3.04 | 4.33 pts | Reported ±1 SE ranges overlap |
| Finance Agent (v2) | 59.04% ±0.27 | 52.25% ±0.35 | 6.79 pts | Reported ±1 SE ranges do not overlap |
| Tax Agent Bench | 57.66% ±3.31 | 65.60% ±3.24 | 7.93 pts | Reported ±1 SE ranges do not overlap |
| MedCode | 53.39% ±2.12 | 49.55% ±2.17 | 3.84 pts | Reported ±1 SE ranges overlap |
| MedScribe | 83.94% ±2.00 | 89.38% ±1.89 | 5.44 pts | Reported ±1 SE ranges do not overlap |
| ProofBench v1.1 | 58.00% ±4.96 | 26.00% ±4.41 | 32.00 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench Science | 5.71% ±2.79 | 11.43% ±3.83 | 5.71 pts | Reported ±1 SE ranges overlap |
| SAGE | 49.23% ±3.38 | 40.79% ±3.35 | 8.44 pts | Reported ±1 SE ranges do not overlap |
| Code Migration | 34.80% ±4.22 | 44.82% ±4.21 | 10.02 pts | Reported ±1 SE ranges do not overlap |
| IOI | 67.83% ±3.61 | 57.72% ±1.99 | 10.11 pts | Reported ±1 SE ranges do not overlap |
| ProgramBench | 0.00% ±0.00 | 0.50% ±0.50 | 0.50 pts | Reported ±1 SE ranges overlap |
| Terminal-Bench 4.0 | 6.06% ±0.88 | 28.28% ±1.34 | 22.22 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 70.39% ±4.84 | 86.17% ±2.18 | 15.78 pts | Reported ±1 SE ranges do not overlap |
| CyberBench v1.1 | 43.75% ±2.21 | 69.46% ±5.67 | 25.71 pts | Reported ±1 SE ranges do not overlap |
Performance by category
| Category | Gemini 3.7 Flash average | Grok 4.7 average |
|---|---|---|
| Index | 38.79% | 40.21% |
| Legal | 43.54% | 48.00% |
| Finance | 62.68% | 61.61% |
| Healthcare | 68.67% | 69.47% |
| Math | 58.00% | 26.00% |
| Science | 5.71% | 11.43% |
| Education | 49.23% | 40.79% |
| Coding | 35.82% | 43.50% |
| Beta | 43.75% | 69.46% |
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 3.7 Flash cost | Grok 4.7 cost | Gemini 3.7 Flash latency | Grok 4.7 latency |
|---|---|---|---|---|
| Vals Index | $4.17 | $11.16 | 43m13s | 32m56s |
| Vals RSI Index | N/A | N/A | 90h00m | 90h00m |
| Harvey's Legal Agent Benchmark | $2.56 | $11.13 | 7m32s | 42m05s |
| Legal Research Bench | $0.76 | $4.92 | 4m49s | 21m14s |
| LegalBench | N/A | N/A | 2.23s | 26.02s |
| EMB | $5.69 | $6.48 | 10m51s | 34m57s |
| Finance Agent (v2) | $1.48 | $2.72 | 3m01s | 15m33s |
| Tax Agent Bench | $0.48 | $1.80 | 87.24s | 16m36s |
| MedCode | N/A | N/A | 9.16s | 3m33s |
| MedScribe | N/A | N/A | 16.82s | 2m33s |
| ProofBench v1.1 | $0.56 | $0.79 | 7m04s | 12m47s |
| Terminal-Bench Science | $13.22 | $21.40 | 1h08m | 2h40m |
| SAGE | N/A | N/A | 9.79s | 4m51s |
| Code Migration | $21.46 | $36.55 | 2h39m | 1h09m |
| IOI | $3.40 | $12.71 | 21m09s | 46m48s |
| ProgramBench | $7.13 | $46.49 | 21m30s | 3h38m |
| Terminal-Bench 4.0 | $14.07 | $14.32 | 1h59m | 1h40m |
| Vibe Code Bench v1.1 | $4.83 | $15.83 | 23m43s | 35m47s |
| CyberBench v1.1 | $2.15 | $8.54 | 7m34s | 15m30s |
Results available only for Gemini 3.7 Flash
- MortgageTax
- TaxEval v2
- GPQA Diamond
- MMLU Pro
- MMMU Pro
- LiveCodeBench
- SkillsBench
- SWE-bench
- SRE Bench
Results available only for Grok 4.7
- BioMysteryBench
- MysteryMechanism
- Public Benefits Bench v1.1