Mercury 2.5 vs Inkling Small: Benchmark Comparison
Mercury 2.5 has the higher score on 1 of 12 shared benchmarks; Inkling Small leads on 10.
The largest observed score gap is 29.02 pts on MedScribe , where Inkling Small leads.
Reported ±1 standard-error ranges overlap on 3 of 12 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 | Mercury 2.5 | Inkling Small | Gap | Reported uncertainty |
|---|---|---|---|---|
| Legal Research Bench | 4.33% ±1.41 | 25.48% ±3.03 | 21.15 pts | Reported ±1 SE ranges do not overlap |
| LegalBench | 83.06% ±0.48 | 82.99% ±0.46 | 0.07 pts | Reported ±1 SE ranges overlap |
| Tax Agent Bench | 2.49% ±1.02 | 15.39% ±2.37 | 12.90 pts | Reported ±1 SE ranges do not overlap |
| MedCode | 31.33% ±1.95 | 37.89% ±2.21 | 6.57 pts | Reported ±1 SE ranges do not overlap |
| MedScribe | 55.09% ±2.10 | 84.11% ±1.87 | 29.02 pts | Reported ±1 SE ranges do not overlap |
| ProofBench v1.1 | 3.00% ±1.71 | 6.00% ±2.39 | 3.00 pts | Reported ±1 SE ranges overlap |
| IOI | 2.44% ±0.68 | 9.33% ±4.47 | 6.89 pts | Reported ±1 SE ranges do not overlap |
| SkillsBench | 18.06% ±3.08 | 33.58% ±3.66 | 15.52 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench 4.0 | 0.00% ±0.00 | 0.00% ±0.00 | 0.00 pts | Reported ±1 SE ranges overlap |
| Vibe Code Bench v1.1 | 3.68% ±1.41 | 19.05% ±3.40 | 15.37 pts | Reported ±1 SE ranges do not overlap |
| CyberBench v1.1 | 45.48% ±6.06 | 64.11% ±6.17 | 18.63 pts | Reported ±1 SE ranges do not overlap |
| Public Benefits Bench v1.1 | 45.26% ±1.29 | 59.41% ±1.28 | 14.14 pts | Reported ±1 SE ranges do not overlap |
Performance by category
| Category | Mercury 2.5 average | Inkling Small average |
|---|---|---|
| Legal | 43.69% | 54.23% |
| Finance | 2.49% | 15.39% |
| Healthcare | 43.21% | 61.00% |
| Math | 3.00% | 6.00% |
| Coding | 6.05% | 15.49% |
| Cyber | 45.48% | 64.11% |
| Social Mobility | 45.26% | 59.41% |
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 | Mercury 2.5 cost | Inkling Small cost | Mercury 2.5 latency | Inkling Small latency |
|---|---|---|---|---|
| Legal Research Bench | $0.06 | $0.14 | 47.61s | 12m55s |
| LegalBench | N/A | N/A | 3.91s | 17.66s |
| Tax Agent Bench | $0.02 | $0.09 | 29.67s | 6m25s |
| MedCode | N/A | N/A | 7.93s | 3m11s |
| MedScribe | N/A | N/A | 9.99s | 3m50s |
| ProofBench v1.1 | $0.04 | $0.08 | 3m22s | 8m27s |
| IOI | $0.16 | $0.34 | 4m03s | 11m18s |
| SkillsBench | $0.10 | $0.15 | 49.36s | 11m41s |
| Terminal-Bench 4.0 | $0.37 | $0.32 | 15m14s | 11m45s |
| Vibe Code Bench v1.1 | $0.38 | $0.43 | 3m13s | 49m09s |
| CyberBench v1.1 | $0.12 | $0.25 | 18m59s | 19m35s |
| Public Benefits Bench v1.1 | $0.07 | $0.10 | 68.33s | 8m41s |
Results available only for Mercury 2.5
- Terminal-Bench-Science
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
- MortgageTax
- TaxEval v2
- GPQA Diamond
- MMLU Pro
- SAGE
- Code Migration v1.1
- LiveCodeBench
- ProgramBench
- SWE-bench