Compare AI models

Specifications
Parameters
7B/13B/33B/65B6B
Context window
2k8k
Benchmarks
HumanEval
23.7%35.9%
Overview
CompanyMetaZ.ai
Release dateFeb 24 2023Jul 25 2023
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

CodeGeeX2-6B leads LLaMA 1 on 1 of the 1 benchmark they both report (HumanEval). LLaMA 1 shipped 151 days before CodeGeeX2-6B, so benchmark comparisons should account for the intervening progress.

LLaMA 1 has 7B/13B/33B/65B parameters, while CodeGeeX2-6B has 6B. Context windows are 2k (LLaMA 1) vs 8k (CodeGeeX2-6B).

On HumanEval, CodeGeeX2-6B leads at 35.9% vs LLaMA 1 at 23.7%.