Claude Sonnet 5.5 vs Grok 4.7: Benchmark Comparison
Claude Sonnet 5.5 has the higher score on 19 of 21 shared benchmarks; Grok 4.7 leads on 2.
The largest observed score gap is 74.00 pts on ProofBench v1.1 , where Claude Sonnet 5.5 leads.
Reported ±1 standard-error ranges overlap on 5 of 21 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 5.5 | Grok 4.7 | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals Index | 67.04% ±0.92 | 54.95% ±1.07 | 12.09 pts | Reported ±1 SE ranges do not overlap |
| Harvey's Legal Agent Benchmark | 2.92% ±1.36 | 12.50% ±2.66 | 9.58 pts | Reported ±1 SE ranges do not overlap |
| Legal Research Bench | 48.08% ±3.47 | 47.12% ±3.47 | 0.96 pts | Reported ±1 SE ranges overlap |
| EMB | 75.71% ±2.44 | 66.99% ±3.04 | 8.72 pts | Reported ±1 SE ranges do not overlap |
| Finance Agent (v2) | 58.10% ±0.67 | 52.25% ±0.35 | 5.85 pts | Reported ±1 SE ranges do not overlap |
| Tax Agent Bench | 41.96% ±3.56 | 34.20% ±3.42 | 7.76 pts | Reported ±1 SE ranges do not overlap |
| MedCode | 52.92% ±2.12 | 49.55% ±2.17 | 3.37 pts | Reported ±1 SE ranges overlap |
| MedScribe | 91.10% ±1.96 | 89.38% ±1.89 | 1.72 pts | Reported ±1 SE ranges overlap |
| ProofBench v1.1 | 100.00% ±0.00 | 26.00% ±4.41 | 74.00 pts | Reported ±1 SE ranges do not overlap |
| BioMysteryBench | 81.11% ±0.64 | 69.26% ±1.33 | 11.85 pts | Reported ±1 SE ranges do not overlap |
| MysteryMechanism | 49.10% ±3.36 | 25.23% ±2.92 | 23.87 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench Science | 45.71% ±6.00 | 10.00% ±3.61 | 35.71 pts | Reported ±1 SE ranges do not overlap |
| SAGE | 51.77% ±3.41 | 40.79% ±3.35 | 10.98 pts | Reported ±1 SE ranges do not overlap |
| Code Migration | 69.83% ±4.26 | 44.82% ±4.21 | 25.01 pts | Reported ±1 SE ranges do not overlap |
| IOI | 83.06% ±3.74 | 57.72% ±1.99 | 25.33 pts | Reported ±1 SE ranges do not overlap |
| ProgramBench | 6.50% ±1.75 | 0.50% ±0.50 | 6.00 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench 4.0 | 64.14% ±1.01 | 28.79% ±2.31 | 35.35 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 92.39% ±1.26 | 86.17% ±2.18 | 6.22 pts | Reported ±1 SE ranges do not overlap |
| CyberBench v1.1 | 59.58% ±5.21 | 69.46% ±5.67 | 9.88 pts | Reported ±1 SE ranges overlap |
| SRE Bench | 30.15% ±2.84 | 3.82% ±1.19 | 26.34 pts | Reported ±1 SE ranges do not overlap |
| Public Benefits Bench v1.1 | 67.19% ±1.22 | 65.63% ±1.24 | 1.56 pts | Reported ±1 SE ranges overlap |
Performance by category
| Category | Claude Sonnet 5.5 average | Grok 4.7 average |
|---|---|---|
| Legal | 25.50% | 29.81% |
| Finance | 58.59% | 51.15% |
| Healthcare | 72.01% | 69.47% |
| Math | 100.00% | 26.00% |
| Science | 58.64% | 34.83% |
| Education | 51.77% | 40.79% |
| Coding | 63.18% | 43.60% |
| Cyber | 44.87% | 36.64% |
| Social Mobility | 67.19% | 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 5.5 cost | Grok 4.7 cost | Claude Sonnet 5.5 latency | Grok 4.7 latency |
|---|---|---|---|---|
| Vals Index | $21.34 | $12.12 | 1h18m | 35m31s |
| Harvey's Legal Agent Benchmark | $16.61 | $11.13 | 57m44s | 42m05s |
| Legal Research Bench | $14.83 | $4.92 | 1h33m | 21m14s |
| EMB | $8.79 | $6.48 | 36m45s | 34m57s |
| Finance Agent (v2) | $6.55 | $2.72 | 31m43s | 15m33s |
| Tax Agent Bench | $9.25 | $1.80 | 59m21s | 16m36s |
| MedCode | N/A | N/A | 3m59s | 3m33s |
| MedScribe | N/A | N/A | 5m07s | 2m33s |
| ProofBench v1.1 | $0.56 | $0.79 | 5m48s | 12m47s |
| BioMysteryBench | $3.74 | $2.49 | 18m06s | 11m37s |
| MysteryMechanism | $3.45 | $1.84 | 23m30s | 17m33s |
| Terminal-Bench Science | $30.43 | $14.45 | 2h39m | 1h29m |
| SAGE | N/A | N/A | 113.42s | 4m51s |
| Code Migration | $75.83 | $36.55 | 3h20m | 1h09m |
| IOI | $7.52 | $12.71 | 45m51s | 46m48s |
| ProgramBench | $35.86 | $46.49 | 2h07m | 3h38m |
| Terminal-Bench 4.0 | $16.51 | $18.09 | 1h22m | 50m21s |
| Vibe Code Bench v1.1 | $31.25 | $15.83 | 1h01m | 35m47s |
| CyberBench v1.1 | $4.42 | $8.54 | 33m53s | 15m30s |
| SRE Bench | $26.52 | $64.77 | 1h55m | 1h41m |
| Public Benefits Bench v1.1 | $4.25 | $2.25 | 1h15m | 16m57s |
Results available only for Claude Sonnet 5.5
None.
Results available only for Grok 4.7
- Vals RSI Index
- LegalBench