Compare AI models

Specifications
Parameters
16B70B
Context window
32k16k
Benchmarks
HumanEval
57.3%53%
Overview
CompanyDeepSeekMeta
Release dateMay 16 2024Jan 29 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

DeepSeek V2 Lite leads Code Llama 70B on 1 of the 1 benchmark they both report (HumanEval). Code Llama 70B shipped 108 days before DeepSeek V2 Lite, so benchmark comparisons should account for the intervening progress.

DeepSeek V2 Lite has 16B parameters, while Code Llama 70B has 70B. Context windows are 32k (DeepSeek V2 Lite) vs 16k (Code Llama 70B).

On HumanEval, DeepSeek V2 Lite leads at 57.3% vs Code Llama 70B at 53%.