Claude Opus 4.8 vs Grok 4.7: Benchmark Comparison
Claude Opus 4.8 has the higher score on 9 of 17 shared benchmarks; Grok 4.7 leads on 8.
The largest observed score gap is 14.00 pts on SAGE , where Claude Opus 4.8 leads.
Reported ±1 standard-error ranges overlap on 12 of 16 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 Opus 4.8 | Grok 4.7 | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals Index | 55.10% ±1.01 | 54.95% ±1.07 | 0.15 pts | Reported ±1 SE ranges overlap |
| Vals RSI Index | 23.98% | 20.20% | 3.78 pts | Uncertainty comparison unavailable |
| Harvey's Legal Agent Benchmark | 9.58% ±2.15 | 12.50% ±2.66 | 2.92 pts | Reported ±1 SE ranges overlap |
| Legal Research Bench | 43.75% ±3.45 | 47.12% ±3.47 | 3.37 pts | Reported ±1 SE ranges overlap |
| LegalBench | 83.57% ±0.47 | 84.39% ±0.46 | 0.82 pts | Reported ±1 SE ranges overlap |
| EMB | 69.37% ±2.61 | 66.99% ±3.04 | 2.38 pts | Reported ±1 SE ranges overlap |
| Finance Agent (v2) | 53.92% ±0.16 | 52.25% ±0.35 | 1.67 pts | Reported ±1 SE ranges do not overlap |
| Tax Agent Bench | 64.91% ±3.17 | 65.60% ±3.24 | 0.69 pts | Reported ±1 SE ranges overlap |
| MedCode | 53.22% ±2.17 | 49.55% ±2.17 | 3.66 pts | Reported ±1 SE ranges overlap |
| MedScribe | 85.75% ±1.93 | 89.38% ±1.89 | 3.62 pts | Reported ±1 SE ranges overlap |
| Terminal-Bench Science | 4.29% ±2.44 | 10.00% ±3.61 | 5.71 pts | Reported ±1 SE ranges overlap |
| SAGE | 54.79% ±3.35 | 40.79% ±3.35 | 14.00 pts | Reported ±1 SE ranges do not overlap |
| Code Migration | 47.25% ±4.18 | 44.82% ±4.21 | 2.43 pts | Reported ±1 SE ranges overlap |
| ProgramBench | 1.00% ±0.70 | 0.50% ±0.50 | 0.50 pts | Reported ±1 SE ranges overlap |
| Terminal-Bench 4.0 | 23.23% ±1.34 | 28.79% ±2.31 | 5.56 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 82.72% ±3.08 | 86.17% ±2.18 | 3.45 pts | Reported ±1 SE ranges overlap |
| Public Benefits Bench v1.1 | 68.13% ±1.21 | 65.63% ±1.24 | 2.50 pts | Reported ±1 SE ranges do not overlap |
Performance by category
| Category | Claude Opus 4.8 average | Grok 4.7 average |
|---|---|---|
| Legal | 45.63% | 48.00% |
| Finance | 62.73% | 61.61% |
| Healthcare | 69.49% | 69.47% |
| Science | 4.29% | 10.00% |
| Education | 54.79% | 40.79% |
| Coding | 38.55% | 40.07% |
| Social Mobility | 68.13% | 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 Opus 4.8 cost | Grok 4.7 cost | Claude Opus 4.8 latency | Grok 4.7 latency |
|---|---|---|---|---|
| Vals Index | $13.14 | $12.12 | 39m48s | 35m31s |
| Vals RSI Index | $465.25 | $237.12 | 90h00m | 90h00m |
| Harvey's Legal Agent Benchmark | $10.22 | $11.13 | 24m08s | 42m05s |
| Legal Research Bench | $2.82 | $4.92 | 11m54s | 21m14s |
| LegalBench | N/A | N/A | 3.11s | 26.02s |
| EMB | $12.06 | $6.48 | 43m52s | 34m57s |
| Finance Agent (v2) | $4.22 | $2.72 | 9m07s | 15m33s |
| Tax Agent Bench | $1.69 | $1.80 | 7m59s | 16m36s |
| MedCode | N/A | N/A | 105.92s | 3m33s |
| MedScribe | N/A | N/A | 82.36s | 2m33s |
| Terminal-Bench Science | $23.14 | $14.45 | 1h55m | 1h29m |
| SAGE | N/A | N/A | 3m06s | 4m51s |
| Code Migration | $30.51 | $36.55 | 1h15m | 1h09m |
| ProgramBench | $31.27 | $46.49 | 1h32m | 3h38m |
| Terminal-Bench 4.0 | $17.14 | $18.09 | 1h11m | 50m21s |
| Vibe Code Bench v1.1 | $26.88 | $15.83 | 1h16m | 35m47s |
| Public Benefits Bench v1.1 | $1.89 | $2.25 | 12m48s | 16m57s |
Results available only for Claude Opus 4.8
- MortgageTax
- TaxEval v2
- GPQA Diamond
- MMLU Pro
- MMMU Pro
- LiveCodeBench
- SkillsBench
- SWE-bench
- Time Horizon Index: KSP
Results available only for Grok 4.7
- ProofBench v1.1
- BioMysteryBench
- MysteryMechanism
- IOI
- CyberBench v1.1