Mercury 2.5 vs Inkling: Benchmark Comparison
Mercury 2.5 has the higher score on 2 of 12 shared benchmarks; Inkling leads on 10.
The largest observed score gap is 30.31 pts on MedScribe , where Inkling leads.
Reported ±1 standard-error ranges overlap on 2 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 | Gap | Reported uncertainty |
|---|---|---|---|---|
| Legal Research Bench | 4.33% ±1.41 | 28.36% ±3.13 | 24.04 pts | Reported ±1 SE ranges do not overlap |
| LegalBench | 83.06% ±0.48 | 82.95% ±0.50 | 0.11 pts | Reported ±1 SE ranges overlap |
| Tax Agent Bench | 2.49% ±1.02 | 15.29% ±2.42 | 12.80 pts | Reported ±1 SE ranges do not overlap |
| MedCode | 31.33% ±1.95 | 41.19% ±2.23 | 9.86 pts | Reported ±1 SE ranges do not overlap |
| MedScribe | 55.09% ±2.10 | 85.41% ±1.84 | 30.31 pts | Reported ±1 SE ranges do not overlap |
| ProofBench v1.1 | 3.00% ±1.71 | 0.00% ±0.00 | 3.00 pts | Reported ±1 SE ranges do not overlap |
| IOI | 2.44% ±0.68 | 14.94% ±5.40 | 12.50 pts | Reported ±1 SE ranges do not overlap |
| SkillsBench | 18.06% ±3.08 | 26.09% ±3.86 | 8.03 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench 4.0 | 0.00% ±0.00 | 0.51% ±0.51 | 0.51 pts | Reported ±1 SE ranges overlap |
| Vibe Code Bench v1.1 | 3.68% ±1.41 | 19.21% ±3.66 | 15.53 pts | Reported ±1 SE ranges do not overlap |
| CyberBench v1.1 | 45.48% ±6.06 | 66.79% ±5.95 | 21.31 pts | Reported ±1 SE ranges do not overlap |
| Public Benefits Bench v1.1 | 45.26% ±1.29 | 58.46% ±1.28 | 13.19 pts | Reported ±1 SE ranges do not overlap |
Performance by category
| Category | Mercury 2.5 average | Inkling average |
|---|---|---|
| Legal | 43.69% | 55.66% |
| Finance | 2.49% | 15.29% |
| Healthcare | 43.21% | 63.30% |
| Math | 3.00% | 0.00% |
| Coding | 6.05% | 15.19% |
| Cyber | 45.48% | 66.79% |
| Social Mobility | 45.26% | 58.46% |
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 cost | Mercury 2.5 latency | Inkling latency |
|---|---|---|---|---|
| Legal Research Bench | $0.06 | $0.49 | 47.61s | 14m29s |
| LegalBench | N/A | N/A | 3.91s | 23.14s |
| Tax Agent Bench | $0.02 | $0.35 | 29.67s | 13m31s |
| MedCode | N/A | N/A | 7.93s | 2m47s |
| MedScribe | N/A | N/A | 9.99s | 4m05s |
| ProofBench v1.1 | $0.04 | $0.20 | 3m22s | 7m16s |
| IOI | $0.16 | $1.71 | 4m03s | 30m12s |
| SkillsBench | $0.10 | $0.83 | 49.36s | 15m39s |
| Terminal-Bench 4.0 | $0.37 | $2.50 | 15m14s | 22m59s |
| Vibe Code Bench v1.1 | $0.38 | $1.61 | 3m13s | 18m25s |
| CyberBench v1.1 | $0.12 | $1.46 | 18m59s | 13m31s |
| Public Benefits Bench v1.1 | $0.07 | $0.37 | 68.33s | 13m31s |
Results available only for Mercury 2.5
- Terminal-Bench-Science
Results available only for Inkling
- 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
- MMMU Pro
- SAGE
- Code Migration v1.1
- LiveCodeBench
- ProgramBench
- SWE-bench
- Vibe Code Bench 1-100
- SRE Bench