Compare AI models

Specifications
Parameters
236B7B/13B/70B
Context window
128k4k
Benchmarks
HumanEval
84.76%29.9%
MATH
71.02%13.5%
Overview
CompanyDeepSeekMeta
Release dateJun 28 2024Jul 18 2023
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

DeepSeek V2-0628 leads LLaMA 2 on 2 of the 2 benchmarks they both report (HumanEval, MATH). LLaMA 2 shipped 346 days before DeepSeek V2-0628, so benchmark comparisons should account for the intervening progress.

DeepSeek V2-0628 has 236B parameters, while LLaMA 2 has 7B/13B/70B. Context windows are 128k (DeepSeek V2-0628) vs 4k (LLaMA 2).

On HumanEval, DeepSeek V2-0628 leads at 84.76% vs LLaMA 2 at 29.9%. On MATH, DeepSeek V2-0628 leads at 71.02% vs LLaMA 2 at 13.5%.