Compare AI models

Specifications
Parameters
671B14B
Context window
128k2k
API pricing
Cheapest input
$0.256DeepInfra—
Cheapest output
$0.712DeepInfra—
$103StreamLake—
Benchmarks
MMLU
88.5%66.3%
Overview
CompanyDeepSeekQwen
Release dateDec 26 2024Sep 25 2023
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

DeepSeek V3 leads Qwen-14B on 1 of the 1 benchmark they both report (MMLU). Qwen-14B shipped 458 days before DeepSeek V3, so benchmark comparisons should account for the intervening progress.

DeepSeek V3 has 671B parameters, while Qwen-14B has 14B. Context windows are 128k (DeepSeek V3) vs 2k (Qwen-14B).

On MMLU, DeepSeek V3 leads at 88.5% vs Qwen-14B at 66.3%.