Gemini 3.6 Flash vs GLM 5.2: Benchmark Comparison
Gemini 3.6 Flash has the higher score on 9 of 15 shared benchmarks; GLM 5.2 leads on 6.
The largest observed score gap is 18.58 pts on LiveCodeBench , where Gemini 3.6 Flash leads.
Reported ±1 standard-error ranges overlap on 6 of 15 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 3.6 Flash | GLM 5.2 | Gap | Reported uncertainty |
|---|---|---|---|---|
| Harvey's Legal Agent Benchmark | 3.33% ±1.17 | 7.08% ±2.00 | 3.75 pts | Reported ±1 SE ranges do not overlap |
| Legal Research Bench | 25.00% ±3.01 | 31.25% ±3.22 | 6.25 pts | Reported ±1 SE ranges do not overlap |
| LegalBench | 86.70% ±0.41 | 84.07% ±0.45 | 2.63 pts | Reported ±1 SE ranges do not overlap |
| EMB | 65.41% ±2.65 | 61.53% ±2.88 | 3.88 pts | Reported ±1 SE ranges overlap |
| Finance Agent (v2) | 56.30% ±0.18 | 49.70% ±0.88 | 6.60 pts | Reported ±1 SE ranges do not overlap |
| TaxEval v2 | 74.86% ±0.85 | 73.34% ±0.86 | 1.51 pts | Reported ±1 SE ranges overlap |
| MedCode | 53.15% ±2.16 | 40.77% ±2.17 | 12.38 pts | Reported ±1 SE ranges do not overlap |
| MedScribe | 79.66% ±1.86 | 83.53% ±2.00 | 3.87 pts | Reported ±1 SE ranges do not overlap |
| GPQA Diamond | 93.43% ±1.33 | 85.61% ±1.77 | 7.83 pts | Reported ±1 SE ranges do not overlap |
| MMLU Pro | 89.28% ±0.30 | 86.71% ±0.33 | 2.56 pts | Reported ±1 SE ranges do not overlap |
| Code Migration | 30.93% ±4.07 | 37.87% ±4.14 | 6.94 pts | Reported ±1 SE ranges overlap |
| LiveCodeBench | 88.08% ±0.94 | 69.50% ±1.18 | 18.58 pts | Reported ±1 SE ranges do not overlap |
| ProgramBench | 0.00% ±0.00 | 0.50% ±0.50 | 0.50 pts | Reported ±1 SE ranges overlap |
| SWE-bench | 79.60% ±1.80 | 82.80% ±1.69 | 3.20 pts | Reported ±1 SE ranges overlap |
| Vibe Code Bench v1.1 | 64.00% ±4.29 | 63.96% ±4.79 | 0.04 pts | Reported ±1 SE ranges overlap |
Performance by category
| Category | Gemini 3.6 Flash average | GLM 5.2 average |
|---|---|---|
| Legal | 38.35% | 40.80% |
| Finance | 65.52% | 61.52% |
| Healthcare | 66.41% | 62.15% |
| Academic | 91.35% | 86.16% |
| Coding | 52.52% | 50.93% |
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 3.6 Flash cost | GLM 5.2 cost | Gemini 3.6 Flash latency | GLM 5.2 latency |
|---|---|---|---|---|
| Harvey's Legal Agent Benchmark | $1.90 | $2.06 | 16m24s | 20m51s |
| Legal Research Bench | $0.47 | $0.89 | 2m48s | 17m03s |
| LegalBench | N/A | N/A | 2.96s | 5.77s |
| EMB | $3.89 | $4.47 | 12m06s | 58m52s |
| Finance Agent (v2) | $1.40 | $0.71 | 3m45s | 7m53s |
| TaxEval v2 | N/A | N/A | 12.34s | 64.94s |
| MedCode | N/A | N/A | 16.68s | 95.27s |
| MedScribe | N/A | N/A | 33.55s | 2m18s |
| GPQA Diamond | N/A | N/A | 18.06s | 3m45s |
| MMLU Pro | N/A | N/A | 7.30s | 51.64s |
| Code Migration | $8.93 | $12.64 | 1h30m | 1h54m |
| LiveCodeBench | N/A | N/A | 31.88s | 8m52s |
| ProgramBench | $5.92 | $12.58 | 47m33s | 2h06m |
| SWE-bench | $1.19 | $0.71 | 13m08s | 11m10s |
| Vibe Code Bench v1.1 | $3.04 | $8.49 | 25m23s | 1h02m |
Results available only for Gemini 3.6 Flash
- MortgageTax
- Tax Agent Bench
- BioMysteryBench
- Terminal-Bench Science
- MMMU Pro
- SAGE
- IOI
- CyberBench v1.1
- Public Benefits Bench v1.1
Results available only for GLM 5.2
- SkillsBench
- SRE Bench