Compare AI models

Specifications
Parameters
236B141B
Context window
128k64k
API pricing
Cheapest input
—$2.00Mistral
Cheapest output
—$6.00Mistral
—$700Mistral · zdr
Benchmarks
HumanEval
84.76%45.1%
Overview
CompanyDeepSeekMistral
Release dateJun 28 2024Apr 10 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

DeepSeek V2-0628 leads Mixtral 8×22B on 1 of the 1 benchmark they both report (HumanEval). Mixtral 8×22B shipped 79 days before DeepSeek V2-0628, so benchmark comparisons should account for the intervening progress.

DeepSeek V2-0628 has 236B parameters, while Mixtral 8×22B has 141B. Context windows are 128k (DeepSeek V2-0628) vs 64k (Mixtral 8×22B).

On HumanEval, DeepSeek V2-0628 leads at 84.76% vs Mixtral 8×22B at 45.1%.