deepseek-coder-6.7B-base-AWQ
This is an AWQ-quantized version of the DeepSeek Coder 6.7B base model, hosted on Hugging Face, primarily useful for efficient local inference.
- Updated
- 2026-06-06
Summary
This is an AWQ-quantized version of the DeepSeek Coder 6.7B base model, hosted on Hugging Face, primarily useful for efficient local inference.
Candidate is sourced from the official 'deepseek-ai/awesome-deepseek-coder' README under 'Quantized Models'.. The Hugging Face page is a model card for an AWQ (Activation-aware Weight Quantization) format, which is a known efficient inference method.. The resource is listed alongside other quantizations (GGUF, GPTQ) for the same base model.