Compare AI models

Specifications
Parameters
7B/13B/34B14.3B
Context window
16k32k
Benchmarks
HumanEval
48.8%34.2%
Overview
CompanyMetaQwen
Release dateAug 24 2023Mar 28 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Code Llama leads Qwen1.5-MoE on 1 of the 1 benchmark they both report (HumanEval). Code Llama shipped 217 days before Qwen1.5-MoE, so benchmark comparisons should account for the intervening progress.

Code Llama has 7B/13B/34B parameters, while Qwen1.5-MoE has 14.3B. Context windows are 16k (Code Llama) vs 32k (Qwen1.5-MoE).

On HumanEval, Code Llama leads at 48.8% vs Qwen1.5-MoE at 34.2%.