DeepSeek MoE 16B
deepseek-moe-16b
Open Source
About
MoE variant with moderate parameter efficiency.
DeepSeek MoE 16B has a 4K-token context window.
Capabilities
VisionMultimodalReasoningFunction CallingTool UseStructured OutputsCode Execution
Specifications
FamilyDeepSeek MoE
Released2024-01-11
Parameters16B
Context4K
ArchitectureMixture of Experts
Specializationgeneral
Trainingfinetuned
Fine-tuningbase