Inkling Small vs MiMo V2.5 Pro: Benchmark Comparison
Inkling Small has the higher score on 10 of 13 shared benchmarks; MiMo V2.5 Pro leads on 3.
The largest observed score gap is 16.00 pts on ProofBench v1.1 , where MiMo V2.5 Pro leads.
Reported ±1 standard-error ranges overlap on 4 of 13 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 | Inkling Small | MiMo V2.5 Pro | Gap | Reported uncertainty |
|---|---|---|---|---|
| Legal Research Bench | 25.48% ±3.03 | 15.87% ±2.54 | 9.62 pts | Reported ±1 SE ranges do not overlap |
| LegalBench | 82.99% ±0.46 | 77.13% ±0.55 | 5.86 pts | Reported ±1 SE ranges do not overlap |
| Tax Agent Bench | 15.39% ±2.37 | 11.50% ±1.96 | 3.89 pts | Reported ±1 SE ranges overlap |
| TaxEval v2 | 75.51% ±0.84 | 73.79% ±0.86 | 1.72 pts | Reported ±1 SE ranges do not overlap |
| MedCode | 37.89% ±2.21 | 32.48% ±1.91 | 5.41 pts | Reported ±1 SE ranges do not overlap |
| MedScribe | 84.11% ±1.87 | 83.73% ±2.06 | 0.38 pts | Reported ±1 SE ranges overlap |
| ProofBench v1.1 | 6.00% ±2.39 | 22.00% ±4.16 | 16.00 pts | Reported ±1 SE ranges do not overlap |
| GPQA Diamond | 83.59% ±2.04 | 82.58% ±2.47 | 1.01 pts | Reported ±1 SE ranges overlap |
| MMLU Pro | 85.57% ±0.35 | 84.59% ±0.45 | 0.97 pts | Reported ±1 SE ranges do not overlap |
| LiveCodeBench | 85.93% ±0.99 | 81.35% ±1.07 | 4.58 pts | Reported ±1 SE ranges do not overlap |
| SWE-bench | 82.20% ±1.71 | 74.00% ±1.96 | 8.20 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 | 19.05% ±3.40 | 34.11% ±4.53 | 15.06 pts | Reported ±1 SE ranges do not overlap |
Performance by category
| Category | Inkling Small average | MiMo V2.5 Pro average |
|---|---|---|
| Legal | 54.23% | 46.50% |
| Finance | 45.45% | 42.65% |
| Healthcare | 61.00% | 58.11% |
| Math | 6.00% | 22.00% |
| Academic | 84.58% | 83.59% |
| Coding | 46.80% | 47.49% |
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 | Inkling Small cost | MiMo V2.5 Pro cost | Inkling Small latency | MiMo V2.5 Pro latency |
|---|---|---|---|---|
| Legal Research Bench | $0.14 | $0.08 | 12m55s | 6m56s |
| LegalBench | N/A | N/A | 17.66s | 27.14s |
| Tax Agent Bench | $0.09 | $0.04 | 6m25s | 6m17s |
| TaxEval v2 | N/A | N/A | 4m02s | 62.83s |
| MedCode | N/A | N/A | 3m11s | 38.45s |
| MedScribe | N/A | N/A | 3m50s | 90.71s |
| ProofBench v1.1 | $0.08 | $0.09 | 8m27s | 19m37s |
| GPQA Diamond | N/A | N/A | 4m20s | 4m11s |
| MMLU Pro | N/A | N/A | 71.69s | 67.76s |
| LiveCodeBench | N/A | N/A | 5m02s | 4m31s |
| SWE-bench | $0.13 | $0.02 | 4m05s | 7m51s |
| Terminal-Bench 4.0 | $0.32 | $0.23 | 11m45s | 2h07m |
| Vibe Code Bench v1.1 | $0.43 | $0.17 | 49m09s | 40m04s |
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
- SAGE
- Code Migration v1.1
- IOI
- ProgramBench
- SkillsBench
- CyberBench v1.1
- Public Benefits Bench v1.1
Results available only for MiMo V2.5 Pro
None.