Compare AI models

Specifications
Parameters
236B7B
Context window
128k64k
Benchmarks
HumanEval
84.76%83.5%
Overview
CompanyDeepSeekQwen
Release dateJun 28 2024Apr 16 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

DeepSeek V2-0628 leads CodeQwen1.5 on 1 of the 1 benchmark they both report (HumanEval). CodeQwen1.5 shipped 73 days before DeepSeek V2-0628, so benchmark comparisons should account for the intervening progress.

DeepSeek V2-0628 has 236B parameters, while CodeQwen1.5 has 7B. Context windows are 128k (DeepSeek V2-0628) vs 64k (CodeQwen1.5).

On HumanEval, DeepSeek V2-0628 leads at 84.76% vs CodeQwen1.5 at 83.5%.