Compare AI models

Specifications
Parameters
16B—
Context window
32k128k
Benchmarks
HumanEval
57.3%92.4%
MMLU
55.7%85.2%
Overview
CompanyDeepSeekOpenAI
Release dateMay 16 2024Sep 12 2024
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

o1-mini leads DeepSeek V2 Lite on 2 of the 2 benchmarks they both report (HumanEval, MMLU). DeepSeek V2 Lite shipped 119 days before o1-mini, so benchmark comparisons should account for the intervening progress.

Context windows are 32k (DeepSeek V2 Lite) vs 128k (o1-mini). DeepSeek V2 Lite is open weight, while o1-mini is closed.

On HumanEval, o1-mini leads at 92.4% vs DeepSeek V2 Lite at 57.3%. On MMLU, o1-mini leads at 85.2% vs DeepSeek V2 Lite at 55.7%.