Compare AI models

Specifications
Parameters
—671B
Context window
100k128k
API pricing
Cheapest input
—$0.256DeepInfra
Cheapest output
—$0.712DeepInfra
—$103StreamLake
Benchmarks
MMLU
78.5%88.5%
Overview
CompanyAnthropicDeepSeek
Release dateJul 11 2023Dec 26 2024
AccessClosedOpen Weight
Model detailsView modelView model

Frequently asked questions

DeepSeek V3 leads Claude 2 on 1 of the 1 benchmark they both report (MMLU). Claude 2 shipped 534 days before DeepSeek V3, so benchmark comparisons should account for the intervening progress.

Context windows are 100k (Claude 2) vs 128k (DeepSeek V3). Claude 2 is closed, while DeepSeek V3 is open weight.

On MMLU, DeepSeek V3 leads at 88.5% vs Claude 2 at 78.5%.