Gemini 3.1 Pro Preview (02/26) vs GLM 5.3 Flash: Benchmark Comparison

Gemini 3.1 Pro Preview (02/26) has the higher score on 7 of 18 shared benchmarks; GLM 5.3 Flash leads on 10.

The largest observed score gap is 34.85 pts on Terminal-Bench 4.0 , where GLM 5.3 Flash leads.

Reported ±1 standard-error ranges overlap on 5 of 18 shared benchmarks with comparable uncertainty data. This is not a pairwise statistical significance test.

Open interactive comparison Read the methodology

Shared benchmark results

Scores and reported standard errors come from the latest visible Vals benchmark version available for both models.

Benchmark Gemini 3.1 Pro Preview (02/26) GLM 5.3 Flash Gap Reported uncertainty
Legal Research Bench 20.67% ±2.81 45.19% ±3.46 24.52 pts Reported ±1 SE ranges do not overlap
LegalBench 87.40% ±0.33 83.93% ±0.43 3.47 pts Reported ±1 SE ranges do not overlap
Tax Agent Bench 9.29% ±2.09 31.55% ±3.35 22.26 pts Reported ±1 SE ranges do not overlap
TaxEval v2 72.88% ±0.86 75.59% ±0.84 2.71 pts Reported ±1 SE ranges do not overlap
MedScribe 76.11% ±1.92 88.94% ±1.91 12.82 pts Reported ±1 SE ranges do not overlap
ProofBench v1.1 26.00% ±4.41 21.00% ±4.09 5.00 pts Reported ±1 SE ranges overlap
Terminal-Bench-Science 1.43% ±1.43 2.86% ±2.01 1.43 pts Reported ±1 SE ranges overlap
GPQA Diamond 95.45% ±1.05 86.36% ±2.13 9.09 pts Reported ±1 SE ranges do not overlap
MMLU Pro 90.99% ±0.28 86.06% ±0.34 4.93 pts Reported ±1 SE ranges do not overlap
MMMU Pro 88.21% ±0.78 86.01% ±0.83 2.20 pts Reported ±1 SE ranges do not overlap
IOI 51.83% ±2.59 52.50% ±2.62 0.67 pts Reported ±1 SE ranges overlap
LiveCodeBench 88.48% ±0.93 80.51% ±1.05 7.97 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
SWE-bench 78.80% ±1.83 92.00% ±1.21 13.20 pts Reported ±1 SE ranges do not overlap
Terminal-Bench 4.0 2.52% ±0.51 37.37% ±2.67 34.85 pts Reported ±1 SE ranges do not overlap
Vibe Code Bench 1-100 6.69% ±2.40 16.03% ±3.28 9.34 pts Reported ±1 SE ranges do not overlap
Vibe Code Bench v1.1 32.03% ±4.34 30.76% ±5.28 1.27 pts Reported ±1 SE ranges overlap
SRE Bench 0.00% ±0.00 0.76% ±0.54 0.76 pts Reported ±1 SE ranges do not overlap

Performance by category

Category Gemini 3.1 Pro Preview (02/26) average GLM 5.3 Flash average
Legal 54.04% 64.56%
Finance 41.09% 53.57%
Healthcare 76.11% 88.94%
Math 26.00% 21.00%
Science 1.43% 2.86%
Academic 91.55% 86.14%
Coding 37.20% 44.17%
Cyber 0.00% 0.76%

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 3.1 Pro Preview (02/26) cost GLM 5.3 Flash cost Gemini 3.1 Pro Preview (02/26) latency GLM 5.3 Flash latency
Legal Research Bench $1.10 $0.12 4m02s 51m07s
LegalBench N/A N/A 10.06s 12.52s
Tax Agent Bench $0.30 $0.14 100.96s 21m37s
TaxEval v2 N/A N/A 41.93s 53.06s
MedScribe N/A N/A 69.14s 86.98s
ProofBench v1.1 $0.78 $0.22 14m49s 1h11m
Terminal-Bench-Science $5.91 $0.52 43m58s 3h28m
GPQA Diamond N/A N/A 65.76s 2m11s
MMLU Pro N/A N/A 23.91s 44.76s
MMMU Pro N/A N/A 76.99s 28.22s
IOI $1.25 $0.43 12m03s 1h32m
LiveCodeBench N/A N/A 88.89s 5m02s
ProgramBench $2.63 $0.80 12m36s 5h30m
SWE-bench $0.78 $0.02 5m12s 1h04m
Terminal-Bench 4.0 $4.11 $1.22 18m44s 1h01m
Vibe Code Bench 1-100 $3.36 $0.46 11m28s 39m03s
Vibe Code Bench v1.1 $3.83 $2.35 20m12s 1h24m
SRE Bench $11.34 $2.29 1h03m 3h15m

Results available only for Gemini 3.1 Pro Preview (02/26)

  • Vals Science Index
  • MortgageTax
  • MedCode
  • BioMysteryBench
  • MysteryMechanism
  • 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
  • SkillsBench
  • CyberBench v1.1
Model details Gemini 3.1 Pro Preview (02/26) Model details GLM 5.3 Flash