Compare AI models

Specifications
Parameters
7B/13B/34B9B
Context window
16k128k
Benchmarks
HumanEval
48.8%82.3%
Overview
CompanyMetaZ.ai
Release dateAug 24 2023Jul 5 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

CodeGeeX4-ALL-9B leads Code Llama on 1 of the 1 benchmark they both report (HumanEval). Code Llama shipped 316 days before CodeGeeX4-ALL-9B, so benchmark comparisons should account for the intervening progress.

Code Llama has 7B/13B/34B parameters, while CodeGeeX4-ALL-9B has 9B. Context windows are 16k (Code Llama) vs 128k (CodeGeeX4-ALL-9B).

On HumanEval, CodeGeeX4-ALL-9B leads at 82.3% vs Code Llama at 48.8%.