Compare AI models

Specifications
Parameters
16B—
Context window
32k200k
Benchmarks
HumanEval
57.3%96.3%
MMLU
55.7%85.9%
Overview
CompanyDeepSeekOpenAI
Release dateMay 16 2024Jan 31 2025
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

o3-mini leads DeepSeek V2 Lite on 2 of the 2 benchmarks they both report (HumanEval, MMLU). DeepSeek V2 Lite shipped 260 days before o3-mini, so benchmark comparisons should account for the intervening progress.

Context windows are 32k (DeepSeek V2 Lite) vs 200k (o3-mini). DeepSeek V2 Lite is open weight, while o3-mini is closed.

On HumanEval, o3-mini leads at 96.3% vs DeepSeek V2 Lite at 57.3%. On MMLU, o3-mini leads at 85.9% vs DeepSeek V2 Lite at 55.7%.