---
language:
- en
---

### OpenLLaMA 3B GGML

OpenLLaMA 3B [350 bt](https://huggingface.co/openlm-research/open_llama_3b_350bt_preview) and [600 bt](https://huggingface.co/openlm-research/open_llama_3b_600bt_preview) converted to [GGML](https://github.com/ggerganov/ggml) format and quantized in q4_0 for inference using [llama.cpp](https://github.com/ggerganov/llama.cpp).