Compare AI models

Specifications
Parameters
1B/4B/12B/27B—
API pricing
Cheapest input
$0.04DeepInfra—
Cheapest output
$0.08DeepInfra—
$15.00DeepInfra · bf16—
Benchmarks
HumanEval
87.8%72%
MATH
89%47.9%
MMLU
76.9%81.5%
GSM8K
95.9%87.6%
Overview
CompanyGoogleZ.ai
Release dateMar 12 2025Jan 16 2024
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

Gemma 3 leads GLM-4 on 3 of the 4 benchmarks they both report (HumanEval, MATH, MMLU, GSM8K). GLM-4 shipped 421 days before Gemma 3, so benchmark comparisons should account for the intervening progress.

Gemma 3 is open weight, while GLM-4 is closed.

On HumanEval, Gemma 3 leads at 87.8% vs GLM-4 at 72%. On MATH, Gemma 3 leads at 89% vs GLM-4 at 47.9%. On MMLU, GLM-4 leads at 81.5% vs Gemma 3 at 76.9%. On GSM8K, Gemma 3 leads at 95.9% vs GLM-4 at 87.6%.