Compare AI models

Specifications
Parameters
16B—
Context window
32k200k
Benchmarks
HumanEval
57.3%97.3%
MMLU
55.7%90%
Overview
CompanyDeepSeekOpenAI
Release dateMay 16 2024Apr 16 2025
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

o4-mini leads DeepSeek V2 Lite on 2 of the 2 benchmarks they both report (HumanEval, MMLU). DeepSeek V2 Lite shipped 335 days before o4-mini, so benchmark comparisons should account for the intervening progress.

Context windows are 32k (DeepSeek V2 Lite) vs 200k (o4-mini). DeepSeek V2 Lite is open weight, while o4-mini is closed.

On HumanEval, o4-mini leads at 97.3% vs DeepSeek V2 Lite at 57.3%. On MMLU, o4-mini leads at 90% vs DeepSeek V2 Lite at 55.7%.