Compare AI models

Specifications
Parameters
7B/13B/34B—
Context window
16k128k
Benchmarks
HumanEval
48.8%75.7%
Overview
CompanyMetaZ.ai
Release dateAug 24 2023Jun 5 2024
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

GLM-4-Air leads Code Llama on 1 of the 1 benchmark they both report (HumanEval). Code Llama shipped 286 days before GLM-4-Air, so benchmark comparisons should account for the intervening progress.

Context windows are 16k (Code Llama) vs 128k (GLM-4-Air). Code Llama is open weight, while GLM-4-Air is closed.

On HumanEval, GLM-4-Air leads at 75.7% vs Code Llama at 48.8%.