Grok 4.5 vs Inkling Small: Benchmark Comparison
Grok 4.5 has the higher score on 16 of 19 shared benchmarks; Inkling Small leads on 3.
The largest observed score gap is 49.95 pts on Vibe Code Bench v1.1 , where Grok 4.5 leads.
Reported ±1 standard-error ranges overlap on 6 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 | Grok 4.5 | Inkling Small | Gap | Reported uncertainty |
|---|---|---|---|---|
| Legal Research Bench | 37.98% ±3.37 | 25.48% ±3.03 | 12.50 pts | Reported ±1 SE ranges do not overlap |
| LegalBench | 85.97% ±0.41 | 82.99% ±0.46 | 2.98 pts | Reported ±1 SE ranges do not overlap |
| MortgageTax | 61.80% ±0.97 | 62.28% ±0.95 | 0.48 pts | Reported ±1 SE ranges overlap |
| Tax Agent Bench | 24.86% ±3.12 | 15.39% ±2.37 | 9.47 pts | Reported ±1 SE ranges do not overlap |
| TaxEval v2 | 71.67% ±0.89 | 75.51% ±0.84 | 3.84 pts | Reported ±1 SE ranges do not overlap |
| MedCode | 43.29% ±2.31 | 37.89% ±2.21 | 5.40 pts | Reported ±1 SE ranges do not overlap |
| MedScribe | 86.88% ±1.94 | 84.11% ±1.87 | 2.77 pts | Reported ±1 SE ranges overlap |
| ProofBench v1.1 | 31.00% ±4.65 | 6.00% ±2.39 | 25.00 pts | Reported ±1 SE ranges do not overlap |
| GPQA Diamond | 92.93% ±1.29 | 83.59% ±2.04 | 9.34 pts | Reported ±1 SE ranges do not overlap |
| MMLU Pro | 89.22% ±0.31 | 85.57% ±0.35 | 3.65 pts | Reported ±1 SE ranges do not overlap |
| SAGE | 34.97% ±3.32 | 32.74% ±3.26 | 2.22 pts | Reported ±1 SE ranges overlap |
| IOI | 40.56% ±3.95 | 9.33% ±4.47 | 31.22 pts | Reported ±1 SE ranges do not overlap |
| LiveCodeBench | 87.35% ±0.96 | 85.93% ±0.99 | 1.42 pts | Reported ±1 SE ranges overlap |
| ProgramBench | 0.00% ±0.00 | 0.50% ±0.50 | 0.50 pts | Reported ±1 SE ranges overlap |
| SkillsBench | 66.03% ±4.60 | 33.58% ±3.66 | 32.45 pts | Reported ±1 SE ranges do not overlap |
| SWE-bench | 86.60% ±1.52 | 82.20% ±1.71 | 4.40 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench 4.0 | 8.59% ±1.34 | 0.00% ±0.00 | 8.59 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 69.00% ±4.48 | 19.05% ±3.40 | 49.95 pts | Reported ±1 SE ranges do not overlap |
| CyberBench v1.1 | 71.01% ±5.79 | 64.11% ±6.17 | 6.91 pts | Reported ±1 SE ranges overlap |
Performance by category
| Category | Grok 4.5 average | Inkling Small average |
|---|---|---|
| Legal | 61.97% | 54.23% |
| Finance | 52.78% | 51.06% |
| Healthcare | 65.09% | 61.00% |
| Math | 31.00% | 6.00% |
| Academic | 91.07% | 84.58% |
| Education | 34.97% | 32.74% |
| Coding | 51.16% | 32.94% |
| Cyber | 71.01% | 64.11% |
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 4.5 cost | Inkling Small cost | Grok 4.5 latency | Inkling Small latency |
|---|---|---|---|---|
| Legal Research Bench | $1.09 | $0.14 | 15m54s | 12m55s |
| LegalBench | N/A | N/A | 67.88s | 17.66s |
| MortgageTax | N/A | N/A | 117.46s | 112.21s |
| Tax Agent Bench | $0.47 | $0.09 | 4m44s | 6m25s |
| TaxEval v2 | N/A | N/A | 3m15s | 4m02s |
| MedCode | N/A | N/A | 59.64s | 3m11s |
| MedScribe | N/A | N/A | 11m45s | 3m50s |
| ProofBench v1.1 | $0.57 | $0.08 | 11m09s | 8m27s |
| GPQA Diamond | N/A | N/A | 3m38s | 4m20s |
| MMLU Pro | N/A | N/A | 5m58s | 71.69s |
| SAGE | N/A | N/A | 2m50s | 3m26s |
| IOI | $5.64 | $0.34 | 1h27m | 11m18s |
| LiveCodeBench | N/A | N/A | 9m49s | 5m02s |
| ProgramBench | $4.12 | $0.48 | 18m59s | 28m31s |
| SkillsBench | $0.58 | $0.15 | 2m16s | 11m41s |
| SWE-bench | $0.54 | $0.13 | 3m20s | 4m05s |
| Terminal-Bench 4.0 | $5.52 | $0.32 | 33m43s | 11m45s |
| Vibe Code Bench v1.1 | $4.10 | $0.43 | 13m50s | 49m09s |
| CyberBench v1.1 | $2.95 | $0.25 | 14m08s | 19m35s |
Results available only for Grok 4.5
- MMMU Pro
- SRE Bench
- Time Horizon Index: KSP
Results available only for Inkling Small
- Vals Finance Index
- Vals Index
- Vals Legal Index
- HLAB-Vals-1.1
- EMB v1.1
- Finance Agent v2.1
- Code Migration v1.1
- Public Benefits Bench v1.1