Compare AI models

Specifications
Parameters
236B—
Context window
128k32k
Benchmarks
HumanEval
87.2%67.7%
Overview
CompanyDeepSeekGoogle
Release dateJul 24 2024Dec 13 2023
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

DeepSeek Coder V2-0724 leads Gemini 1.0 Pro on 1 of the 1 benchmark they both report (HumanEval). Gemini 1.0 Pro shipped 224 days before DeepSeek Coder V2-0724, so benchmark comparisons should account for the intervening progress.

Context windows are 128k (DeepSeek Coder V2-0724) vs 32k (Gemini 1.0 Pro). DeepSeek Coder V2-0724 is open weight, while Gemini 1.0 Pro is closed.

On HumanEval, DeepSeek Coder V2-0724 leads at 87.2% vs Gemini 1.0 Pro at 67.7%.