Model comparison
Claude Sonnet 5 vs Grok 4.7: Benchmark Comparison
Claude Sonnet 5 has the higher score on 4 of 19 shared benchmarks; Grok 4.7 leads on 15.
The largest observed score gap is 51.00 pts on ProofBench v1.1 , where Claude Sonnet 5 leads.
Reported ±1 standard-error ranges overlap on 10 of 19 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 | Grok 4.7 | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals Index | 51.77% ±1.09 | 54.95% ±1.07 | 3.17 pts | Reported ±1 SE ranges do not overlap |
| Harvey's Legal Agent Benchmark | 5.00% ±1.65 | 12.50% ±2.42 | 7.50 pts | Reported ±1 SE ranges do not overlap |
| Legal Research Bench | 41.83% ±3.43 | 47.12% ±3.47 | 5.29 pts | Reported ±1 SE ranges overlap |
| LegalBench | 83.92% ±0.46 | 84.39% ±0.46 | 0.47 pts | Reported ±1 SE ranges overlap |
| EMB | 66.32% ±3.01 | 66.99% ±3.04 | 0.67 pts | Reported ±1 SE ranges overlap |
| Finance Agent (v2) | 53.91% ±0.52 | 52.25% ±0.35 | 1.66 pts | Reported ±1 SE ranges do not overlap |
| Tax Agent Bench | 62.27% ±3.19 | 65.60% ±3.24 | 3.32 pts | Reported ±1 SE ranges overlap |
| MedCode | 47.54% ±2.27 | 49.55% ±2.17 | 2.01 pts | Reported ±1 SE ranges overlap |
| MedScribe | 76.05% ±3.05 | 89.38% ±1.89 | 13.32 pts | Reported ±1 SE ranges do not overlap |
| ProofBench v1.1 | 77.00% ±4.23 | 26.00% ±4.41 | 51.00 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench Science | 2.86% ±2.01 | 10.00% ±3.61 | 7.14 pts | Reported ±1 SE ranges do not overlap |
| SAGE | 48.92% ±3.40 | 40.79% ±3.35 | 8.13 pts | Reported ±1 SE ranges do not overlap |
| Code Migration | 44.39% ±4.25 | 44.82% ±4.21 | 0.42 pts | Reported ±1 SE ranges overlap |
| IOI | 45.00% ±2.75 | 57.72% ±1.99 | 12.72 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 | 9.60% ±1.01 | 28.79% ±2.31 | 19.19 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 81.33% ±3.05 | 86.17% ±2.18 | 4.85 pts | Reported ±1 SE ranges overlap |
| CyberBench v1.1 | 61.91% ±5.74 | 69.46% ±5.67 | 7.56 pts | Reported ±1 SE ranges overlap |
| Public Benefits Bench v1.1 | 66.03% ±1.23 | 65.63% ±1.24 | 0.41 pts | Reported ±1 SE ranges overlap |
Performance by category
| Category | Claude Sonnet 5 average | Grok 4.7 average |
|---|---|---|
| Legal | 43.58% | 48.00% |
| Finance | 60.83% | 61.61% |
| Healthcare | 61.80% | 69.47% |
| Math | 77.00% | 26.00% |
| Science | 2.86% | 10.00% |
| Education | 48.92% | 40.79% |
| Coding | 36.06% | 43.60% |
| Beta | 61.91% | 69.46% |
| Social Mobility | 66.03% | 65.63% |
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 cost | Grok 4.7 cost | Claude Sonnet 5 latency | Grok 4.7 latency |
|---|---|---|---|---|
| Vals Index | $13.72 | $12.12 | 54m16s | 35m31s |
| Harvey's Legal Agent Benchmark | $8.95 | $11.13 | 38m52s | 42m05s |
| Legal Research Bench | $2.72 | $4.92 | 25m46s | 21m14s |
| LegalBench | N/A | N/A | 4.90s | 26.02s |
| EMB | $10.29 | $6.48 | 49m46s | 34m57s |
| Finance Agent (v2) | $0.75 | $2.72 | 13m12s | 15m33s |
| Tax Agent Bench | $1.79 | $1.80 | 19m19s | 16m36s |
| MedCode | N/A | N/A | 2m15s | 3m33s |
| MedScribe | N/A | N/A | 4m12s | 2m33s |
| ProofBench v1.1 | $1.37 | $0.79 | 14m58s | 12m47s |
| Terminal-Bench Science | $18.70 | $14.45 | 2h55m | 1h29m |
| SAGE | N/A | N/A | 7m14s | 4m51s |
| Code Migration | $35.31 | $36.55 | 1h57m | 1h09m |
| IOI | $12.89 | $12.71 | 59m44s | 46m48s |
| ProgramBench | $24.36 | $46.49 | 1h31m | 3h38m |
| Terminal-Bench 4.0 | $26.33 | $18.09 | 1h45m | 50m21s |
| Vibe Code Bench v1.1 | $25.39 | $15.83 | 1h07m | 35m47s |
| CyberBench v1.1 | $1.89 | $8.54 | 16m57s | 15m30s |
| Public Benefits Bench v1.1 | $1.29 | $2.25 | 21m09s | 16m57s |
Results available only for Claude Sonnet 5
- MortgageTax
- TaxEval v2
- GPQA Diamond
- MMLU Pro
- MMMU Pro
- LiveCodeBench
- SkillsBench
- SWE-bench
- Vibe Code Bench 1-100
Results available only for Grok 4.7
- Vals RSI Index
- BioMysteryBench
- MysteryMechanism