Compare AI models

Specifications
Context window
100k128k
Benchmarks
HumanEval
71.2%75.7%
MMLU
78.5%81.9%
GSM8K
88%90.9%
Overview
CompanyAnthropicZ.ai
Release dateJul 11 2023Jun 5 2024
AccessClosedClosed
Model detailsView modelView model

Frequently asked questions

GLM-4-Air leads Claude 2 on 3 of the 3 benchmarks they both report (HumanEval, MMLU, GSM8K). Claude 2 shipped 330 days before GLM-4-Air, so benchmark comparisons should account for the intervening progress.

Context windows are 100k (Claude 2) vs 128k (GLM-4-Air).

On HumanEval, GLM-4-Air leads at 75.7% vs Claude 2 at 71.2%. On MMLU, GLM-4-Air leads at 81.9% vs Claude 2 at 78.5%. On GSM8K, GLM-4-Air leads at 90.9% vs Claude 2 at 88%.