| license: apache-2.0 | |
| ## Introduce | |
| Quantizing the [UnicomLLM/Unichat-llama3-Chinese-8B](https://huggingface.co/UnicomLLM/Unichat-llama3-Chinese-8B) to f16, q2, q3, q4, q5, q6 and q8 with Llama.cpp. | |
| ## Prompt template | |
| ``` | |
| {system_message} | |
| Human: {prompt} | |
| Assistant: | |
| ``` |