Compare AI models

Specifications
Context window
32k128k
Benchmarks
HumanEval
74.4%75.7%
MATH
53.2%57.9%
MMLU
83.7%81.9%
GSM8K
88.9%90.9%
Overview
CompanyGoogleZ.ai
Release dateFeb 8 2024Jun 5 2024
AccessClosedClosed
Model detailsView modelView model

Frequently asked questions

GLM-4-Air leads Gemini 1.0 Ultra on 3 of the 4 benchmarks they both report (HumanEval, MATH, MMLU, GSM8K). Gemini 1.0 Ultra shipped 118 days before GLM-4-Air, so benchmark comparisons should account for the intervening progress.

Context windows are 32k (Gemini 1.0 Ultra) vs 128k (GLM-4-Air).

On HumanEval, GLM-4-Air leads at 75.7% vs Gemini 1.0 Ultra at 74.4%. On MATH, GLM-4-Air leads at 57.9% vs Gemini 1.0 Ultra at 53.2%. On MMLU, Gemini 1.0 Ultra leads at 83.7% vs GLM-4-Air at 81.9%. On GSM8K, GLM-4-Air leads at 90.9% vs Gemini 1.0 Ultra at 88.9%.