Compare AI models

Specifications
Parameters
—16B
Context window
200k32k
Benchmarks
HumanEval
88.1%57.3%
MMLU
80.9%55.7%
Overview
CompanyAnthropicDeepSeek
Release dateOct 22 2024May 16 2024
AccessClosedOpen Weight
Model detailsView modelView model

Frequently asked questions

Claude 3.5 Haiku leads DeepSeek V2 Lite on 2 of the 2 benchmarks they both report (HumanEval, MMLU). DeepSeek V2 Lite shipped 159 days before Claude 3.5 Haiku, so benchmark comparisons should account for the intervening progress.

Context windows are 200k (Claude 3.5 Haiku) vs 32k (DeepSeek V2 Lite). Claude 3.5 Haiku is closed, while DeepSeek V2 Lite is open weight.

On HumanEval, Claude 3.5 Haiku leads at 88.1% vs DeepSeek V2 Lite at 57.3%. On MMLU, Claude 3.5 Haiku leads at 80.9% vs DeepSeek V2 Lite at 55.7%.