Compare AI models

Specifications
Parameters
671B—
Context window
128k200k
API pricing
Cheapest input
$0.256DeepInfra—
Cheapest output
$0.712DeepInfra—
$103StreamLake—
Benchmarks
SWE-Bench Verified
42%49.3%
GPQA Diamond
59.1%77.2%
MMLU
88.5%86.9%
Overview
CompanyDeepSeekOpenAI
Release dateDec 26 2024Jan 31 2025
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

o3-mini-high leads DeepSeek V3 on 2 of the 3 benchmarks they both report (SWE-Bench Verified, GPQA Diamond, MMLU). DeepSeek V3 shipped 36 days before o3-mini-high, so benchmark comparisons should account for the intervening progress.

Context windows are 128k (DeepSeek V3) vs 200k (o3-mini-high). DeepSeek V3 is open weight, while o3-mini-high is closed.

On SWE-Bench Verified, o3-mini-high leads at 49.3% vs DeepSeek V3 at 42%. On GPQA Diamond, o3-mini-high leads at 77.2% vs DeepSeek V3 at 59.1%. On MMLU, DeepSeek V3 leads at 88.5% vs o3-mini-high at 86.9%.