Compare AI models

Specifications
Parameters
236B70B
Context window
128k16k
Benchmarks
HumanEval
84.76%53%
Overview
CompanyDeepSeekMeta
Release dateJun 28 2024Jan 29 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

DeepSeek V2-0628 leads Code Llama 70B on 1 of the 1 benchmark they both report (HumanEval). Code Llama 70B shipped 151 days before DeepSeek V2-0628, so benchmark comparisons should account for the intervening progress.

DeepSeek V2-0628 has 236B parameters, while Code Llama 70B has 70B. Context windows are 128k (DeepSeek V2-0628) vs 16k (Code Llama 70B).

On HumanEval, DeepSeek V2-0628 leads at 84.76% vs Code Llama 70B at 53%.