Compare AI models

Specifications
Parameters
7B/13B/34B16B
Context window
16k8k
Benchmarks
HumanEval
48.8%48.1%
Overview
CompanyMetaMoonshot AI
Release dateAug 24 2023Feb 22 2025
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Code Llama leads Moonlight on 1 of the 1 benchmark they both report (HumanEval). Code Llama shipped 548 days before Moonlight, so benchmark comparisons should account for the intervening progress.

Code Llama has 7B/13B/34B parameters, while Moonlight has 16B. Context windows are 16k (Code Llama) vs 8k (Moonlight).

On HumanEval, Code Llama leads at 48.8% vs Moonlight at 48.1%.