Compare AI models

Specifications
Parameters
236B—
Context window
128k128k
API pricing
Input price
—$0.15
Output price
—$0.60
Cached input price
—$0.075
Cheapest input
—$0.15Azure
Cheapest output
—$0.60Azure
—$60.00OpenAI list price
Benchmarks
HumanEval
84.76%87.2%
MATH
71.02%70.2%
Overview
CompanyDeepSeekOpenAI
Release dateJun 28 2024Jul 18 2024
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

DeepSeek V2-0628 and GPT-4o mini are evenly matched across the 2 benchmarks they both report (HumanEval, MATH). Only GPT-4o mini has a verified first-party price for an available API: $0.15 per million input tokens and $0.60 per million output tokens. No currently available pay-as-you-go API rate is tracked for DeepSeek V2-0628. DeepSeek V2-0628 shipped 20 days before GPT-4o mini, so benchmark comparisons should account for the intervening progress.

Context windows are 128k (DeepSeek V2-0628) vs 128k (GPT-4o mini). DeepSeek V2-0628 is open weight, while GPT-4o mini is closed.

On HumanEval, GPT-4o mini leads at 87.2% vs DeepSeek V2-0628 at 84.76%. On MATH, DeepSeek V2-0628 leads at 71.02% vs GPT-4o mini at 70.2%.