Compare AI models

Specifications
Parameters
236B—
Context window
128k—
Benchmarks
HumanEval
81.1%85.7%
MATH
53.9%73%
MMLU
77.8%86.2%
Overview
CompanyDeepSeekSpaceXSI
Release dateMay 6 2024Aug 13 2024
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

Grok‑2 mini leads DeepSeek V2 on 3 of the 3 benchmarks they both report (HumanEval, MATH, MMLU). DeepSeek V2 shipped 99 days before Grok‑2 mini, so benchmark comparisons should account for the intervening progress.

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

On HumanEval, Grok‑2 mini leads at 85.7% vs DeepSeek V2 at 81.1%. On MATH, Grok‑2 mini leads at 73% vs DeepSeek V2 at 53.9%. On MMLU, Grok‑2 mini leads at 86.2% vs DeepSeek V2 at 77.8%.