Quantization made by Richard Erkhov.

[Github](https://github.com/RichardErkhov)

[Discord](https://discord.gg/pvy7H8DZMG)

[Request more models](https://github.com/RichardErkhov/quant_request)


ShieldLM-14B-qwen - GGUF
- Model creator: https://huggingface.co/thu-coai/
- Original model: https://huggingface.co/thu-coai/ShieldLM-14B-qwen/


| Name | Quant method | Size |
| ---- | ---- | ---- |
| [ShieldLM-14B-qwen.Q2_K.gguf](https://huggingface.co/RichardErkhov/thu-coai_-_ShieldLM-14B-qwen-gguf/blob/main/ShieldLM-14B-qwen.Q2_K.gguf) | Q2_K | 5.41GB |
| [ShieldLM-14B-qwen.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/thu-coai_-_ShieldLM-14B-qwen-gguf/blob/main/ShieldLM-14B-qwen.IQ3_XS.gguf) | IQ3_XS | 6.12GB |
| [ShieldLM-14B-qwen.IQ3_S.gguf](https://huggingface.co/RichardErkhov/thu-coai_-_ShieldLM-14B-qwen-gguf/blob/main/ShieldLM-14B-qwen.IQ3_S.gguf) | IQ3_S | 6.31GB |
| [ShieldLM-14B-qwen.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/thu-coai_-_ShieldLM-14B-qwen-gguf/blob/main/ShieldLM-14B-qwen.Q3_K_S.gguf) | Q3_K_S | 6.31GB |
| [ShieldLM-14B-qwen.IQ3_M.gguf](https://huggingface.co/RichardErkhov/thu-coai_-_ShieldLM-14B-qwen-gguf/blob/main/ShieldLM-14B-qwen.IQ3_M.gguf) | IQ3_M | 6.87GB |
| [ShieldLM-14B-qwen.Q3_K.gguf](https://huggingface.co/RichardErkhov/thu-coai_-_ShieldLM-14B-qwen-gguf/blob/main/ShieldLM-14B-qwen.Q3_K.gguf) | Q3_K | 7.16GB |
| [ShieldLM-14B-qwen.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/thu-coai_-_ShieldLM-14B-qwen-gguf/blob/main/ShieldLM-14B-qwen.Q3_K_M.gguf) | Q3_K_M | 7.16GB |
| [ShieldLM-14B-qwen.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/thu-coai_-_ShieldLM-14B-qwen-gguf/blob/main/ShieldLM-14B-qwen.Q3_K_L.gguf) | Q3_K_L | 7.44GB |
| [ShieldLM-14B-qwen.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/thu-coai_-_ShieldLM-14B-qwen-gguf/blob/main/ShieldLM-14B-qwen.IQ4_XS.gguf) | IQ4_XS | 7.37GB |
| [ShieldLM-14B-qwen.Q4_0.gguf](https://huggingface.co/RichardErkhov/thu-coai_-_ShieldLM-14B-qwen-gguf/blob/main/ShieldLM-14B-qwen.Q4_0.gguf) | Q4_0 | 7.62GB |
| [ShieldLM-14B-qwen.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/thu-coai_-_ShieldLM-14B-qwen-gguf/blob/main/ShieldLM-14B-qwen.IQ4_NL.gguf) | IQ4_NL | 7.68GB |
| [ShieldLM-14B-qwen.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/thu-coai_-_ShieldLM-14B-qwen-gguf/blob/main/ShieldLM-14B-qwen.Q4_K_S.gguf) | Q4_K_S | 7.96GB |
| [ShieldLM-14B-qwen.Q4_K.gguf](https://huggingface.co/RichardErkhov/thu-coai_-_ShieldLM-14B-qwen-gguf/blob/main/ShieldLM-14B-qwen.Q4_K.gguf) | Q4_K | 8.8GB |
| [ShieldLM-14B-qwen.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/thu-coai_-_ShieldLM-14B-qwen-gguf/blob/main/ShieldLM-14B-qwen.Q4_K_M.gguf) | Q4_K_M | 8.8GB |
| [ShieldLM-14B-qwen.Q4_1.gguf](https://huggingface.co/RichardErkhov/thu-coai_-_ShieldLM-14B-qwen-gguf/blob/main/ShieldLM-14B-qwen.Q4_1.gguf) | Q4_1 | 8.4GB |
| [ShieldLM-14B-qwen.Q5_0.gguf](https://huggingface.co/RichardErkhov/thu-coai_-_ShieldLM-14B-qwen-gguf/blob/main/ShieldLM-14B-qwen.Q5_0.gguf) | Q5_0 | 9.18GB |
| [ShieldLM-14B-qwen.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/thu-coai_-_ShieldLM-14B-qwen-gguf/blob/main/ShieldLM-14B-qwen.Q5_K_S.gguf) | Q5_K_S | 9.34GB |
| [ShieldLM-14B-qwen.Q5_K.gguf](https://huggingface.co/RichardErkhov/thu-coai_-_ShieldLM-14B-qwen-gguf/blob/main/ShieldLM-14B-qwen.Q5_K.gguf) | Q5_K | 10.14GB |
| [ShieldLM-14B-qwen.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/thu-coai_-_ShieldLM-14B-qwen-gguf/blob/main/ShieldLM-14B-qwen.Q5_K_M.gguf) | Q5_K_M | 10.14GB |
| [ShieldLM-14B-qwen.Q5_1.gguf](https://huggingface.co/RichardErkhov/thu-coai_-_ShieldLM-14B-qwen-gguf/blob/main/ShieldLM-14B-qwen.Q5_1.gguf) | Q5_1 | 9.96GB |
| [ShieldLM-14B-qwen.Q6_K.gguf](https://huggingface.co/RichardErkhov/thu-coai_-_ShieldLM-14B-qwen-gguf/blob/main/ShieldLM-14B-qwen.Q6_K.gguf) | Q6_K | 11.46GB |
| [ShieldLM-14B-qwen.Q8_0.gguf](https://huggingface.co/RichardErkhov/thu-coai_-_ShieldLM-14B-qwen-gguf/blob/main/ShieldLM-14B-qwen.Q8_0.gguf) | Q8_0 | 14.03GB |




Original model description:
---
license: mit
language:
- en
- zh
---
## Introduction
The ShieldLM model ([paper link](https://arxiv.org/abs/2402.16444)) initialized from [Qwen-14B-Chat](https://huggingface.co/Qwen/Qwen-14B-Chat). ShieldLM is a bilingual (Chinese and English) safety detector that mainly aims to help to detect safety issues in LLMs' generations. It aligns with general human safety standards, supports fine-grained customizable detection rules, and provides explanations for its decisions. 
Refer to our [github repository](https://github.com/thu-coai/ShieldLM) for more detailed information.

## Usage
Please refer to our [github repository](https://github.com/thu-coai/ShieldLM) for the detailed usage instructions.

## Performance
ShieldLM demonstrates impressive detection performance across 4 ID and OOD test sets, compared to strong baselines such as GPT-4, Llama Guard and Perspective API.
Refer to [our paper](https://arxiv.org/abs/2402.16444) for more detailed evaluation results.