Gemini 3.5 Flash vs GLM 5.3 Flash: Benchmark Comparison
Gemini 3.5 Flash has the higher score on 8 of 16 shared benchmarks; GLM 5.3 Flash leads on 7.
The largest observed score gap is 31.31 pts on Terminal-Bench 4.0 , where GLM 5.3 Flash leads.
Reported ±1 standard-error ranges overlap on 4 of 16 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 | Gemini 3.5 Flash | GLM 5.3 Flash | Gap | Reported uncertainty |
|---|---|---|---|---|
| Legal Research Bench | 30.77% ±3.21 | 45.19% ±3.46 | 14.42 pts | Reported ±1 SE ranges do not overlap |
| LegalBench | 83.60% ±0.85 | 83.93% ±0.43 | 0.33 pts | Reported ±1 SE ranges overlap |
| Tax Agent Bench | 21.65% ±2.96 | 31.55% ±3.35 | 9.90 pts | Reported ±1 SE ranges do not overlap |
| TaxEval v2 | 74.37% ±0.85 | 75.59% ±0.84 | 1.23 pts | Reported ±1 SE ranges overlap |
| MedScribe | 76.57% ±1.92 | 88.94% ±1.91 | 12.36 pts | Reported ±1 SE ranges do not overlap |
| ProofBench v1.1 | 31.00% ±4.65 | 21.00% ±4.09 | 10.00 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench-Science | 5.71% ±2.79 | 2.86% ±2.01 | 2.86 pts | Reported ±1 SE ranges overlap |
| GPQA Diamond | 92.68% ±1.46 | 86.36% ±2.13 | 6.31 pts | Reported ±1 SE ranges do not overlap |
| MMLU Pro | 89.52% ±0.31 | 86.06% ±0.34 | 3.46 pts | Reported ±1 SE ranges do not overlap |
| MMMU Pro | 88.27% ±0.77 | 86.01% ±0.83 | 2.25 pts | Reported ±1 SE ranges do not overlap |
| LiveCodeBench | 87.60% ±0.95 | 80.51% ±1.05 | 7.09 pts | Reported ±1 SE ranges do not overlap |
| ProgramBench | 0.00% ±0.00 | 0.00% ±0.00 | 0.00 pts | Reported ±1 SE ranges overlap |
| SkillsBench | 52.74% ±4.47 | 40.16% ±4.43 | 12.58 pts | Reported ±1 SE ranges do not overlap |
| SWE-bench | 78.80% ±1.83 | 92.00% ±1.21 | 13.20 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench 4.0 | 6.06% ±1.51 | 37.37% ±2.67 | 31.31 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 48.68% ±4.73 | 30.76% ±5.28 | 17.92 pts | Reported ±1 SE ranges do not overlap |
Performance by category
| Category | Gemini 3.5 Flash average | GLM 5.3 Flash average |
|---|---|---|
| Legal | 57.19% | 64.56% |
| Finance | 48.01% | 53.57% |
| Healthcare | 76.57% | 88.94% |
| Math | 31.00% | 21.00% |
| Science | 5.71% | 2.86% |
| Academic | 90.15% | 86.14% |
| Coding | 45.65% | 46.80% |
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 | Gemini 3.5 Flash cost | GLM 5.3 Flash cost | Gemini 3.5 Flash latency | GLM 5.3 Flash latency |
|---|---|---|---|---|
| Legal Research Bench | $1.31 | $0.12 | 4m38s | 51m07s |
| LegalBench | N/A | N/A | 34.58s | 12.52s |
| Tax Agent Bench | $0.86 | $0.14 | 3m53s | 21m37s |
| TaxEval v2 | N/A | N/A | 19.69s | 53.06s |
| MedScribe | N/A | N/A | 57.08s | 86.98s |
| ProofBench v1.1 | $0.57 | $0.22 | 6m34s | 1h11m |
| Terminal-Bench-Science | $3.80 | $0.52 | 1h01m | 3h28m |
| GPQA Diamond | N/A | N/A | 27.66s | 2m11s |
| MMLU Pro | N/A | N/A | 26.26s | 44.76s |
| MMMU Pro | N/A | N/A | 12.23s | 28.22s |
| LiveCodeBench | N/A | N/A | 64.37s | 5m02s |
| ProgramBench | N/A | N/A | 18m25s | 5h30m |
| SkillsBench | $1.25 | $0.04 | 7m15s | 12m09s |
| SWE-bench | $0.95 | $0.02 | 4m14s | 1h04m |
| Terminal-Bench 4.0 | $5.98 | $1.22 | 1h11m | 1h01m |
| Vibe Code Bench v1.1 | $2.54 | $2.35 | 14m53s | 1h24m |
Results available only for Gemini 3.5 Flash
- MortgageTax
- MedCode
- SAGE
- Public Benefits Bench v1.1
Results available only for GLM 5.3 Flash
- Vals Finance Index
- Vals Index
- Vals Legal Index
- HLAB-Vals-1.1
- EMB v1.1
- Finance Agent v2.1
- Code Migration v1.1
- IOI
- Vibe Code Bench 1-100
- CyberBench v1.1
- SRE Bench