Compare AI models

Specifications
Parameters
—236B
Context window
200k128k
Benchmarks
HumanEval
75.9%84.76%
Overview
CompanyAnthropicDeepSeek
Release dateMar 4 2024Jun 28 2024
AccessClosedOpen Weight
Model detailsView modelView model

Frequently asked questions

DeepSeek V2-0628 leads Claude 3 Haiku on 1 of the 1 benchmark they both report (HumanEval). Claude 3 Haiku shipped 116 days before DeepSeek V2-0628, so benchmark comparisons should account for the intervening progress.

Context windows are 200k (Claude 3 Haiku) vs 128k (DeepSeek V2-0628). Claude 3 Haiku is closed, while DeepSeek V2-0628 is open weight.

On HumanEval, DeepSeek V2-0628 leads at 84.76% vs Claude 3 Haiku at 75.9%.