Compare AI models

Specifications
Parameters
671B—
Context window
128k128k
API pricing
Cheapest input
$0.256DeepInfra—
Cheapest output
$0.712DeepInfra—
$103StreamLake—
Benchmarks
MMLU
88.5%87.5%
Overview
CompanyDeepSeekSpaceXSI
Release dateDec 26 2024Aug 13 2024
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

DeepSeek V3 leads Grok‑2 on 1 of the 1 benchmark they both report (MMLU). Grok‑2 shipped 135 days before DeepSeek V3, so benchmark comparisons should account for the intervening progress.

Context windows are 128k (DeepSeek V3) vs 128k (Grok‑2). DeepSeek V3 is open weight, while Grok‑2 is closed.

On MMLU, DeepSeek V3 leads at 88.5% vs Grok‑2 at 87.5%.