Compare AI models

Specifications
Parameters
671B—
Context window
128k400k
API pricing
Input price
—$0.05
Output price
—$0.40
Cached input price
—$0.005
Cheapest input
$0.256DeepInfra$0.025OpenAI
Cheapest output
$0.712DeepInfra$0.20OpenAI
$103StreamLake$30.00OpenAI list price
Benchmarks
SWE-Bench Verified
42%54.7%
GPQA Diamond
59.1%71.2%
Overview
CompanyDeepSeekOpenAI
Release dateDec 26 2024Aug 7 2025
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

GPT-5 nano leads DeepSeek V3 on 2 of the 2 benchmarks they both report (SWE-Bench Verified, GPQA Diamond). Only GPT-5 nano has a verified first-party price for an available API: $0.05 per million input tokens and $0.40 per million output tokens. No currently available pay-as-you-go API rate is tracked for DeepSeek V3. DeepSeek V3 shipped 224 days before GPT-5 nano, so benchmark comparisons should account for the intervening progress.

Context windows are 128k (DeepSeek V3) vs 400k (GPT-5 nano). DeepSeek V3 is open weight, while GPT-5 nano is closed.

On SWE-Bench Verified, GPT-5 nano leads at 54.7% vs DeepSeek V3 at 42%. On GPQA Diamond, GPT-5 nano leads at 71.2% vs DeepSeek V3 at 59.1%.