Compare AI models

Specifications
Parameters
9B/27B—
Context window
8k—
API pricing
Cheapest input
$0.65NextBit—
Cheapest output
$0.65NextBit—
$163NextBit · int4—
Benchmarks
HumanEval
51.8%72%
MATH
42.3%47.9%
MMLU
75.2%81.5%
GSM8K
74%87.6%
Overview
CompanyGoogleZ.ai
Release dateJun 27 2024Jan 16 2024
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

GLM-4 leads Gemma 2 on 4 of the 4 benchmarks they both report (HumanEval, MATH, MMLU, GSM8K). GLM-4 shipped 163 days before Gemma 2, so benchmark comparisons should account for the intervening progress.

Gemma 2 is open weight, while GLM-4 is closed.

On HumanEval, GLM-4 leads at 72% vs Gemma 2 at 51.8%. On MATH, GLM-4 leads at 47.9% vs Gemma 2 at 42.3%. On MMLU, GLM-4 leads at 81.5% vs Gemma 2 at 75.2%. On GSM8K, GLM-4 leads at 87.6% vs Gemma 2 at 74%.