Compare AI models

Specifications
Parameters
—16B
Context window
100k32k
Benchmarks
HumanEval
58.7%57.3%
GSM8K
86.7%72%
Overview
CompanyAnthropicDeepSeek
Release dateAug 9 2023May 16 2024
AccessClosedOpen Weight
Model detailsView modelView model

Frequently asked questions

Claude Instant 1.2 leads DeepSeek V2 Lite on 2 of the 2 benchmarks they both report (HumanEval, GSM8K). Claude Instant 1.2 shipped 281 days before DeepSeek V2 Lite, so benchmark comparisons should account for the intervening progress.

Context windows are 100k (Claude Instant 1.2) vs 32k (DeepSeek V2 Lite). Claude Instant 1.2 is closed, while DeepSeek V2 Lite is open weight.

On HumanEval, Claude Instant 1.2 leads at 58.7% vs DeepSeek V2 Lite at 57.3%. On GSM8K, Claude Instant 1.2 leads at 86.7% vs DeepSeek V2 Lite at 72%.