Compare AI models

Specifications
Parameters
671B—
Context window
128k—
Benchmarks
HumanEval
65.2%85.7%
MMLU
87.1%86.2%
Overview
CompanyDeepSeekSpaceXSI
Release dateDec 25 2024Aug 13 2024
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

DeepSeek V3 Base and Grok‑2 mini are evenly matched across the 2 benchmarks they both report (HumanEval, MMLU). Grok‑2 mini shipped 134 days before DeepSeek V3 Base, so benchmark comparisons should account for the intervening progress.

DeepSeek V3 Base is open weight, while Grok‑2 mini is closed.

On HumanEval, Grok‑2 mini leads at 85.7% vs DeepSeek V3 Base at 65.2%. On MMLU, DeepSeek V3 Base leads at 87.1% vs Grok‑2 mini at 86.2%.