second-state
/

Qwen2.5-3B-Instruct-GGUF

@@ -1,38 +1,87 @@
-# FACET
-## How to run
-```bash
-python main.py --full-name <hf-account-name/hf-model-name> -s <target-directory-to-save-git-clone> --enable-converter -c <path-to-python-convert-script> --enable-quantizer -q <path-to-llamacpp-quantizer> -t q4_0
-```
-For example,
-```bash
-python main.py --full-name baichuan-inc/Baichuan2-13B-Chat -s /home/ubuntu/workspace/models/ --enable-converter -c /home/ubuntu/workspace/llama.cpp/convert.py --enable-quantizer -q llama-cpp-quantizer -t Q5_K_M
-```
-## Allowed quantization types
-```text
-   2  or  Q4_0   :  3.56G, +0.2166 ppl @ LLaMA-v1-7B
-   3  or  Q4_1   :  3.90G, +0.1585 ppl @ LLaMA-v1-7B
-   8  or  Q5_0   :  4.33G, +0.0683 ppl @ LLaMA-v1-7B
-   9  or  Q5_1   :  4.70G, +0.0349 ppl @ LLaMA-v1-7B
-  10  or  Q2_K   :  2.63G, +0.6717 ppl @ LLaMA-v1-7B
-  12  or  Q3_K   : alias for Q3_K_M
-  11  or  Q3_K_S :  2.75G, +0.5551 ppl @ LLaMA-v1-7B
-  12  or  Q3_K_M :  3.07G, +0.2496 ppl @ LLaMA-v1-7B
-  13  or  Q3_K_L :  3.35G, +0.1764 ppl @ LLaMA-v1-7B
-  15  or  Q4_K   : alias for Q4_K_M
-  14  or  Q4_K_S :  3.59G, +0.0992 ppl @ LLaMA-v1-7B
-  15  or  Q4_K_M :  3.80G, +0.0532 ppl @ LLaMA-v1-7B
-  17  or  Q5_K   : alias for Q5_K_M
-  16  or  Q5_K_S :  4.33G, +0.0400 ppl @ LLaMA-v1-7B
-  17  or  Q5_K_M :  4.45G, +0.0122 ppl @ LLaMA-v1-7B
-  18  or  Q6_K   :  5.15G, -0.0008 ppl @ LLaMA-v1-7B
-   7  or  Q8_0   :  6.70G, +0.0004 ppl @ LLaMA-v1-7B
-   1  or  F16    : 13.00G              @ 7B
-   0  or  F32    : 26.00G              @ 7B
-          COPY   : only copy tensors, no quantizing
-```

+---
+base_model: Qwen/Qwen2.5-3B-Instruct
+license: other
+license_name: qwen-research
+license_link: https://huggingface.co/Qwen/Qwen2.5-3B-Instruct/blob/main/LICENSEE
+model_creator: Qwen
+model_name: QQwen2.5-3B-Instruct
+quantized_by: Second State Inc.
+language:
+- en
+pipeline_tag: text-generation
+tags:
+- chat
+---
+<!-- header start -->
+<!-- 200823 -->
+<div style="width: auto; margin-left: auto; margin-right: auto">
+<img src="https://github.com/LlamaEdge/LlamaEdge/raw/dev/assets/logo.svg" style="width: 100%; min-width: 400px; display: block; margin: auto;">
+</div>
+<hr style="margin-top: 1.0em; margin-bottom: 1.0em;">
+<!-- header end -->
+# Qwen2.5-3B-Instruct-GGUF
+## Original Model
+[Qwen/Qwen2.5-3B-Instruct](https://huggingface.co/Qwen/Qwen2.5-3B-Instruct)
+## Run with LlamaEdge
+- LlamaEdge version: [v0.14.3](https://github.com/LlamaEdge/LlamaEdge/releases/tag/0.14.3) and above
+- Prompt template
+  - Prompt type: `chatml`
+  - Prompt string
+    ```text
+    <|im_start|>system
+    {system_message}<|im_end|>
+    <|im_start|>user
+    {prompt}<|im_end|>
+    <|im_start|>assistant
+    ```
+- Context size: `32000`
+- Run as LlamaEdge service
+  ```bash
+  wasmedge --dir .:. --nn-preload default:GGML:AUTO:QQwen2.5-3B-Instruct-Q5_K_M.gguf \
+    llama-api-server.wasm \
+    --model-name QQwen2.5-3B-Instruct \
+    --prompt-template chatml \
+    --ctx-size 32000
+  ```
+- Run as LlamaEdge command app
+  ```bash
+  wasmedge --dir .:. --nn-preload default:GGML:AUTO:QQwen2.5-3B-Instruct-Q5_K_M.gguf \
+    llama-chat.wasm \
+    --prompt-template chatml \
+    --ctx-size 32000
+  ```
+## Quantized GGUF Models
+| Name | Quant method | Bits | Size | Use case |
+| ---- | ---- | ---- | ---- | ----- |
+| [QQwen2.5-3B-Instruct-Q2_K.gguf](https://huggingface.co/second-state/QQwen2.5-3B-Instruct-GGUF/blob/main/QQwen2.5-3B-Instruct-Q2_K.gguf)     | Q2_K   | 2 | 676 MB| smallest, significant quality loss - not recommended for most purposes |
+| [QQwen2.5-3B-Instruct-Q3_K_L.gguf](https://huggingface.co/second-state/QQwen2.5-3B-Instruct-GGUF/blob/main/QQwen2.5-3B-Instruct-Q3_K_L.gguf) | Q3_K_L | 3 | 880 MB| small, substantial quality loss |
+| [QQwen2.5-3B-Instruct-Q3_K_M.gguf](https://huggingface.co/second-state/QQwen2.5-3B-Instruct-GGUF/blob/main/QQwen2.5-3B-Instruct-Q3_K_M.gguf) | Q3_K_M | 3 | 824 MB| very small, high quality loss |
+| [QQwen2.5-3B-Instruct-Q3_K_S.gguf](https://huggingface.co/second-state/QQwen2.5-3B-Instruct-GGUF/blob/main/QQwen2.5-3B-Instruct-Q3_K_S.gguf) | Q3_K_S | 3 | 761 MB| very small, high quality loss |
+| [QQwen2.5-3B-Instruct-Q4_0.gguf](https://huggingface.co/second-state/QQwen2.5-3B-Instruct-GGUF/blob/main/QQwen2.5-3B-Instruct-Q4_0.gguf)     | Q4_0   | 4 | 935 MB| legacy; small, very high quality loss - prefer using Q3_K_M |
+| [QQwen2.5-3B-Instruct-Q4_K_M.gguf](https://huggingface.co/second-state/QQwen2.5-3B-Instruct-GGUF/blob/main/QQwen2.5-3B-Instruct-Q4_K_M.gguf) | Q4_K_M | 4 | 986 MB| medium, balanced quality - recommended |
+| [QQwen2.5-3B-Instruct-Q4_K_S.gguf](https://huggingface.co/second-state/QQwen2.5-3B-Instruct-GGUF/blob/main/QQwen2.5-3B-Instruct-Q4_K_S.gguf) | Q4_K_S | 4 | 940 MB| small, greater quality loss |
+| [QQwen2.5-3B-Instruct-Q5_0.gguf](https://huggingface.co/second-state/QQwen2.5-3B-Instruct-GGUF/blob/main/QQwen2.5-3B-Instruct-Q5_0.gguf)     | Q5_0   | 5 | 1.1 GB| legacy; medium, balanced quality - prefer using Q4_K_M |
+| [QQwen2.5-3B-Instruct-Q5_K_M.gguf](https://huggingface.co/second-state/QQwen2.5-3B-Instruct-GGUF/blob/main/QQwen2.5-3B-Instruct-Q5_K_M.gguf) | Q5_K_M | 5 | 1.13 GB| large, very low quality loss - recommended |
+| [QQwen2.5-3B-Instruct-Q5_K_S.gguf](https://huggingface.co/second-state/QQwen2.5-3B-Instruct-GGUF/blob/main/QQwen2.5-3B-Instruct-Q5_K_S.gguf) | Q5_K_S | 5 | 1.1 GB| large, low quality loss - recommended |
+| [QQwen2.5-3B-Instruct-Q6_K.gguf](https://huggingface.co/second-state/QQwen2.5-3B-Instruct-GGUF/blob/main/QQwen2.5-3B-Instruct-Q6_K.gguf)     | Q6_K   | 6 | 1.27 GB| very large, extremely low quality loss |
+| [QQwen2.5-3B-Instruct-Q8_0.gguf](https://huggingface.co/second-state/QQwen2.5-3B-Instruct-GGUF/blob/main/QQwen2.5-3B-Instruct-Q8_0.gguf)     | Q8_0   | 8 | 1.65 GB| very large, extremely low quality loss - not recommended |
+| [QQwen2.5-3B-Instruct-f16.gguf](https://huggingface.co/second-state/QQwen2.5-3B-Instruct-GGUF/blob/main/QQwen2.5-3B-Instruct-f16.gguf)       | f16    | 16 | 3.09 GB| |
+*Quantized with llama.cpp b3751*