Compare AI models

Specifications
Parameters
671B6B
Context window
128k32k
API pricing
Cheapest input
$0.256DeepInfra—
Cheapest output
$0.712DeepInfra—
$103StreamLake—
Benchmarks
MMLU
88.5%45.2%
Overview
CompanyDeepSeekZ.ai
Release dateDec 26 2024Jun 25 2023
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

DeepSeek V3 leads ChatGLM2-6B on 1 of the 1 benchmark they both report (MMLU). ChatGLM2-6B shipped 550 days before DeepSeek V3, so benchmark comparisons should account for the intervening progress.

DeepSeek V3 has 671B parameters, while ChatGLM2-6B has 6B. Context windows are 128k (DeepSeek V3) vs 32k (ChatGLM2-6B).

On MMLU, DeepSeek V3 leads at 88.5% vs ChatGLM2-6B at 45.2%.