Grok 4.6 vs GPT 5.5: Benchmark Comparison
Grok 4.6 has the higher score on 12 of 22 shared benchmarks; GPT 5.5 leads on 10.
The largest observed score gap is 22.63 pts on SAGE , where GPT 5.5 leads.
Reported ±1 standard-error ranges overlap on 9 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 | Grok 4.6 | GPT 5.5 | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals RSI Index | 25.46% | 16.48% | 8.98 pts | Uncertainty comparison unavailable |
| Harvey's Legal Agent Benchmark | 15.83% ±2.65 | 3.75% ±1.17 | 12.08 pts | Reported ±1 SE ranges do not overlap |
| Legal Research Bench | 48.08% ±3.47 | 40.38% ±3.41 | 7.69 pts | Reported ±1 SE ranges do not overlap |
| LegalBench | 86.31% ±0.42 | 86.52% ±0.41 | 0.21 pts | Reported ±1 SE ranges overlap |
| EMB | 62.73% ±3.08 | 64.54% ±2.87 | 1.82 pts | Reported ±1 SE ranges overlap |
| Finance Agent (v2) | 53.68% ±0.67 | 51.76% ±0.55 | 1.92 pts | Reported ±1 SE ranges do not overlap |
| MortgageTax | 64.19% ±0.95 | 68.76% ±0.91 | 4.57 pts | Reported ±1 SE ranges do not overlap |
| Tax Agent Bench | 70.79% ±3.00 | 60.46% ±3.22 | 10.33 pts | Reported ±1 SE ranges do not overlap |
| TaxEval v2 | 71.10% ±0.90 | 74.98% ±0.85 | 3.88 pts | Reported ±1 SE ranges do not overlap |
| MedCode | 44.71% ±2.26 | 49.10% ±2.19 | 4.39 pts | Reported ±1 SE ranges overlap |
| MedScribe | 86.53% ±1.96 | 86.87% ±1.93 | 0.33 pts | Reported ±1 SE ranges overlap |
| GPQA Diamond | 94.70% ±1.13 | 93.18% ±1.29 | 1.52 pts | Reported ±1 SE ranges overlap |
| MMLU Pro | 89.40% ±0.30 | 88.14% ±0.32 | 1.26 pts | Reported ±1 SE ranges do not overlap |
| SAGE | 28.90% ±3.08 | 51.53% ±3.95 | 22.63 pts | Reported ±1 SE ranges do not overlap |
| Code Migration | 44.57% ±4.48 | 45.16% ±4.16 | 0.59 pts | Reported ±1 SE ranges overlap |
| LiveCodeBench | 88.22% ±0.94 | 85.30% ±1.02 | 2.93 pts | Reported ±1 SE ranges do not overlap |
| ProgramBench | 1.00% ±0.70 | 0.50% ±0.50 | 0.50 pts | Reported ±1 SE ranges overlap |
| SkillsBench | 55.77% ±4.84 | 62.21% ±4.52 | 6.44 pts | Reported ±1 SE ranges overlap |
| SWE-bench | 95.60% ±0.92 | 82.60% ±1.70 | 13.00 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 76.24% ±3.82 | 69.85% ±4.54 | 6.39 pts | Reported ±1 SE ranges overlap |
| Public Benefits Bench v1.1 | 66.85% ±1.23 | 60.89% ±1.27 | 5.95 pts | Reported ±1 SE ranges do not overlap |
| Time Horizon Index: KSP | 5.83% ±0.00 | 8.33% ±0.00 | 2.50 pts | Reported ±1 SE ranges do not overlap |
Performance by category
| Category | Grok 4.6 average | GPT 5.5 average |
|---|---|---|
| Legal | 50.07% | 43.55% |
| Finance | 64.50% | 64.10% |
| Healthcare | 65.62% | 67.98% |
| Academic | 92.05% | 90.66% |
| Education | 28.90% | 51.53% |
| Coding | 60.23% | 57.60% |
| Social Mobility | 66.85% | 60.89% |
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 | Grok 4.6 cost | GPT 5.5 cost | Grok 4.6 latency | GPT 5.5 latency |
|---|---|---|---|---|
| Vals RSI Index | $88.12 | $857.16 | 90h00m | 90h00m |
| Harvey's Legal Agent Benchmark | $4.01 | $4.60 | 45m05s | 12m14s |
| Legal Research Bench | $1.53 | $7.40 | 25m04s | 37m45s |
| LegalBench | N/A | N/A | 27.42s | 18.14s |
| EMB | $3.06 | $3.27 | 52m59s | 15m11s |
| Finance Agent (v2) | $1.66 | $4.15 | 16m41s | 11m02s |
| MortgageTax | N/A | N/A | 8.55s | 28.24s |
| Tax Agent Bench | $0.98 | $4.07 | 10m58s | 15m52s |
| TaxEval v2 | N/A | N/A | 47.57s | 95.83s |
| MedCode | N/A | N/A | 2m07s | 2m40s |
| MedScribe | N/A | N/A | 70.94s | 2m13s |
| GPQA Diamond | N/A | N/A | 3m29s | 112.33s |
| MMLU Pro | N/A | N/A | 54.98s | 42.11s |
| SAGE | N/A | N/A | 3m22s | 76.05s |
| Code Migration | $14.58 | $6.44 | 1h30m | 31m35s |
| LiveCodeBench | N/A | N/A | 118.40s | 2m47s |
| ProgramBench | $16.76 | $6.95 | 4h27m | 22m03s |
| SkillsBench | $0.92 | $2.54 | 13m28s | 6m55s |
| SWE-bench | $0.78 | $1.36 | 10m04s | 7m06s |
| Vibe Code Bench v1.1 | $4.88 | $16.66 | 25m30s | 31m52s |
| Public Benefits Bench v1.1 | $0.92 | $3.97 | 25m53s | 39m37s |
| Time Horizon Index: KSP | $225.12 | $516.36 | N/A | N/A |
Results available only for Grok 4.6
- Vals Index
- ProofBench v1.1
- BioMysteryBench
- MysteryMechanism
- Terminal-Bench Science
- IOI
- Terminal-Bench 4.0
- Vibe Code Bench 1-100
- CyberBench v1.1
Results available only for GPT 5.5
- MMMU Pro
- SRE Bench