Compare AI models

Specifications
Parameters
16B/236B—
Context window
128k—
Benchmarks
HumanEval
90.2%85.7%
MATH
75.7%73%
MMLU
79.2%86.2%
Overview
CompanyDeepSeekSpaceXSI
Release dateJun 14 2024Aug 13 2024
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

DeepSeek Coder V2 leads Grok‑2 mini on 2 of the 3 benchmarks they both report (HumanEval, MATH, MMLU). DeepSeek Coder V2 shipped 60 days before Grok‑2 mini, so benchmark comparisons should account for the intervening progress.

DeepSeek Coder V2 is open weight, while Grok‑2 mini is closed.

On HumanEval, DeepSeek Coder V2 leads at 90.2% vs Grok‑2 mini at 85.7%. On MATH, DeepSeek Coder V2 leads at 75.7% vs Grok‑2 mini at 73%. On MMLU, Grok‑2 mini leads at 86.2% vs DeepSeek Coder V2 at 79.2%.