Compare AI models

Specifications
Parameters
16B7B/13B/34B
Context window
32k16k
Benchmarks
HumanEval
57.3%48.8%
Overview
CompanyDeepSeekMeta
Release dateMay 16 2024Aug 24 2023
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

DeepSeek V2 Lite leads Code Llama on 1 of the 1 benchmark they both report (HumanEval). Code Llama shipped 266 days before DeepSeek V2 Lite, so benchmark comparisons should account for the intervening progress.

DeepSeek V2 Lite has 16B parameters, while Code Llama has 7B/13B/34B. Context windows are 32k (DeepSeek V2 Lite) vs 16k (Code Llama).

On HumanEval, DeepSeek V2 Lite leads at 57.3% vs Code Llama at 48.8%.