Compare AI models

Specifications
Parameters
236B8B/70B
Context window
128k8k
Benchmarks
HumanEval
87.2%81.7%
Overview
CompanyDeepSeekMeta
Release dateJul 24 2024Apr 18 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

DeepSeek Coder V2-0724 leads LLaMA 3 (8B/70B) on 1 of the 1 benchmark they both report (HumanEval). LLaMA 3 (8B/70B) shipped 97 days before DeepSeek Coder V2-0724, so benchmark comparisons should account for the intervening progress.

DeepSeek Coder V2-0724 has 236B parameters, while LLaMA 3 (8B/70B) has 8B/70B. Context windows are 128k (DeepSeek Coder V2-0724) vs 8k (LLaMA 3 (8B/70B)).

On HumanEval, DeepSeek Coder V2-0724 leads at 87.2% vs LLaMA 3 (8B/70B) at 81.7%.