deepseek-moe-16b-chat-W4A16-GPTQ

Description

deepseek-moe-16b-chat-W4A16-GPTQ is an open-weight text-generation model with roughly 16B parameters published by soyrsoyr on the Hugging Face Hub. It has 210 downloads. Runs with transformers.

Categories

• Coding
• Productivity
• Design

Tags

• deepseek
• text-generation
• moe
• gptq
• quantized
• compressed-tensors
• vllm
• llm-compressor

Capabilities

• Text generation
• Conversational
• Library: transformers
• 16B parameters