Inkling Small vs GLM 5.3 Flash: Benchmark Comparison
Inkling Small has the higher score on 2 of 23 shared benchmarks; GLM 5.3 Flash leads on 21.
The largest observed score gap is 43.17 pts on IOI , where GLM 5.3 Flash leads.
Reported ±1 standard-error ranges overlap on 6 of 23 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 | GLM 5.3 Flash | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals Finance Index | 36.27% ±1.56 | 55.83% ±1.51 | 19.56 pts | Reported ±1 SE ranges do not overlap |
| Vals Index | 27.59% ±1.13 | 50.86% ±1.25 | 23.27 pts | Reported ±1 SE ranges do not overlap |
| Vals Legal Index | 57.83% ±1.70 | 71.05% ±1.83 | 13.22 pts | Reported ±1 SE ranges do not overlap |
| HLAB-Vals-1.1 | 2.92% ±1.36 | 10.42% ±2.57 | 7.50 pts | Reported ±1 SE ranges do not overlap |
| Legal Research Bench | 25.48% ±3.03 | 45.19% ±3.46 | 19.71 pts | Reported ±1 SE ranges do not overlap |
| LegalBench | 82.99% ±0.46 | 83.93% ±0.43 | 0.94 pts | Reported ±1 SE ranges do not overlap |
| EMB v1.1 | 40.82% ±3.21 | 66.25% ±2.69 | 25.43 pts | Reported ±1 SE ranges do not overlap |
| Finance Agent v2.1 | 44.78% ±2.10 | 60.59% ±2.03 | 15.81 pts | Reported ±1 SE ranges do not overlap |
| Tax Agent Bench | 15.39% ±2.37 | 31.55% ±3.35 | 16.16 pts | Reported ±1 SE ranges do not overlap |
| TaxEval v2 | 75.51% ±0.84 | 75.59% ±0.84 | 0.08 pts | Reported ±1 SE ranges overlap |
| MedScribe | 84.11% ±1.87 | 88.94% ±1.91 | 4.82 pts | Reported ±1 SE ranges do not overlap |
| ProofBench v1.1 | 6.00% ±2.39 | 21.00% ±4.09 | 15.00 pts | Reported ±1 SE ranges do not overlap |
| GPQA Diamond | 83.59% ±2.04 | 86.36% ±2.13 | 2.78 pts | Reported ±1 SE ranges overlap |
| MMLU Pro | 85.57% ±0.35 | 86.06% ±0.34 | 0.49 pts | Reported ±1 SE ranges overlap |
| Code Migration v1.1 | 10.43% ±2.32 | 51.65% ±2.99 | 41.22 pts | Reported ±1 SE ranges do not overlap |
| IOI | 9.33% ±4.47 | 52.50% ±2.62 | 43.17 pts | Reported ±1 SE ranges do not overlap |
| LiveCodeBench | 85.93% ±0.99 | 80.51% ±1.05 | 5.42 pts | Reported ±1 SE ranges do not overlap |
| ProgramBench | 0.50% ±0.50 | 0.00% ±0.00 | 0.50 pts | Reported ±1 SE ranges overlap |
| SkillsBench | 33.58% ±3.66 | 40.16% ±4.43 | 6.58 pts | Reported ±1 SE ranges overlap |
| SWE-bench | 82.20% ±1.71 | 92.00% ±1.21 | 9.80 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench 4.0 | 0.00% ±0.00 | 37.37% ±2.67 | 37.37 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 19.05% ±3.40 | 30.76% ±5.28 | 11.71 pts | Reported ±1 SE ranges do not overlap |
| CyberBench v1.1 | 64.11% ±6.17 | 67.86% ±5.57 | 3.75 pts | Reported ±1 SE ranges overlap |
Performance by category
| Category | Inkling Small average | GLM 5.3 Flash average |
|---|---|---|
| Legal | 37.13% | 46.51% |
| Finance | 44.12% | 58.50% |
| Healthcare | 84.11% | 88.94% |
| Math | 6.00% | 21.00% |
| Academic | 84.58% | 86.21% |
| Coding | 30.13% | 48.12% |
| Cyber | 64.11% | 67.86% |
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 | GLM 5.3 Flash cost | Inkling Small latency | GLM 5.3 Flash latency |
|---|---|---|---|---|
| Vals Finance Index | $0.18 | $0.31 | 10m24s | 33m49s |
| Vals Index | $0.40 | $1.13 | 16m26s | 1h04m |
| Vals Legal Index | $0.14 | $0.19 | 8m58s | 28m23s |
| HLAB-Vals-1.1 | $0.18 | $0.32 | 7m35s | 12m25s |
| Legal Research Bench | $0.14 | $0.12 | 12m55s | 51m07s |
| LegalBench | N/A | N/A | 17.66s | 12.52s |
| EMB v1.1 | $0.25 | $0.71 | 19m34s | 1h15m |
| Finance Agent v2.1 | $0.21 | $0.07 | 5m14s | 4m56s |
| Tax Agent Bench | $0.09 | $0.14 | 6m25s | 21m37s |
| TaxEval v2 | N/A | N/A | 4m02s | 53.06s |
| MedScribe | N/A | N/A | 3m50s | 86.98s |
| ProofBench v1.1 | $0.08 | $0.22 | 8m27s | 1h11m |
| GPQA Diamond | N/A | N/A | 4m20s | 2m11s |
| MMLU Pro | N/A | N/A | 71.69s | 44.76s |
| Code Migration v1.1 | $1.36 | $4.48 | 17m56s | 3h20m |
| IOI | $0.34 | $0.43 | 11m18s | 1h32m |
| LiveCodeBench | N/A | N/A | 5m02s | 5m02s |
| ProgramBench | $0.48 | $0.80 | 28m31s | 5h30m |
| SkillsBench | $0.15 | $0.04 | 11m41s | 12m09s |
| SWE-bench | $0.13 | $0.02 | 4m05s | 1h04m |
| Terminal-Bench 4.0 | $0.32 | $1.22 | 11m45s | 1h01m |
| Vibe Code Bench v1.1 | $0.43 | $2.35 | 49m09s | 1h24m |
| CyberBench v1.1 | $0.25 | $0.14 | 19m35s | 35m48s |
Results available only for Inkling Small
- MortgageTax
- MedCode
- SAGE
- Public Benefits Bench v1.1
Results available only for GLM 5.3 Flash
- Terminal-Bench-Science
- MMMU Pro
- Vibe Code Bench 1-100
- SRE Bench