second-state
/

Qwen3-14B-GGUF

@@ -1,88 +1,109 @@
----
-base_model: Qwen/Qwen3-14B
-license: apache-2.0
-license_link: https://huggingface.co/Qwen/Qwen3-14B/blob/main/LICENSE
-model_creator: Qwen
-model_name: Qwen3-14B
-quantized_by: Second State Inc.
-language:
-- en
-pipeline_tag: text-generation
-tags:
-- chat
-library_name: transformers
----
-<!-- 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 -->
-# Qwen3-14B-GGUF
-## Original Model
-[Qwen/Qwen3-14B](https://huggingface.co/Qwen/Qwen3-14B)
-## Run with LlamaEdge
-<!-- - LlamaEdge version: coming soon -->
-- LlamaEdge version: [v0.17.0](https://github.com/LlamaEdge/LlamaEdge/releases/tag/0.17.0) 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: `128000`
-- Run as LlamaEdge service
-  ```bash
-  wasmedge --dir .:. --nn-preload default:GGML:AUTO:Qwen3-14B-Q5_K_M.gguf \
-    llama-api-server.wasm \
-    --model-name Qwen3-14B \
-    --prompt-template chatml \
-    --ctx-size 128000
-  ```
-- Run as LlamaEdge command app
-  ```bash
-  wasmedge --dir .:. --nn-preload default:GGML:AUTO:Qwen3-14B-Q5_K_M.gguf \
-    llama-chat.wasm \
-    --prompt-template chatml \
-    --ctx-size 128000
-  ```
-## Quantized GGUF Models
-| Name | Quant method | Bits | Size | Use case |
-| ---- | ---- | ---- | ---- | ----- |
-| [Qwen3-14B-Q2_K.gguf](https://huggingface.co/second-state/Qwen3-14B-GGUF/blob/main/Qwen3-14B-Q2_K.gguf)     | Q2_K   | 2 | 5.75 GB| smallest, significant quality loss - not recommended for most purposes |
-| [Qwen3-14B-Q3_K_L.gguf](https://huggingface.co/second-state/Qwen3-14B-GGUF/blob/main/Qwen3-14B-Q3_K_L.gguf) | Q3_K_L | 3 | 7.90 GB| small, substantial quality loss |
-| [Qwen3-14B-Q3_K_M.gguf](https://huggingface.co/second-state/Qwen3-14B-GGUF/blob/main/Qwen3-14B-Q3_K_M.gguf) | Q3_K_M | 3 | 7.32 GB| very small, high quality loss |
-| [Qwen3-14B-Q3_K_S.gguf](https://huggingface.co/second-state/Qwen3-14B-GGUF/blob/main/Qwen3-14B-Q3_K_S.gguf) | Q3_K_S | 3 | 6.66 GB| very small, high quality loss |
-| [Qwen3-14B-Q4_0.gguf](https://huggingface.co/second-state/Qwen3-14B-GGUF/blob/main/Qwen3-14B-Q4_0.gguf)     | Q4_0   | 4 | 8.52 GB| legacy; small, very high quality loss - prefer using Q3_K_M |
-| [Qwen3-14B-Q4_K_M.gguf](https://huggingface.co/second-state/Qwen3-14B-GGUF/blob/main/Qwen3-14B-Q4_K_M.gguf) | Q4_K_M | 4 | 9.00 GB| medium, balanced quality - recommended |
-| [Qwen3-14B-Q4_K_S.gguf](https://huggingface.co/second-state/Qwen3-14B-GGUF/blob/main/Qwen3-14B-Q4_K_S.gguf) | Q4_K_S | 4 | 8.57 GB| small, greater quality loss |
-| [Qwen3-14B-Q5_0.gguf](https://huggingface.co/second-state/Qwen3-14B-GGUF/blob/main/Qwen3-14B-Q5_0.gguf)     | Q5_0   | 5 | 10.3 GB| legacy; medium, balanced quality - prefer using Q4_K_M |
-| [Qwen3-14B-Q5_K_M.gguf](https://huggingface.co/second-state/Qwen3-14B-GGUF/blob/main/Qwen3-14B-Q5_K_M.gguf) | Q5_K_M | 5 | 10.5 GB| large, very low quality loss - recommended |
-| [Qwen3-14B-Q5_K_S.gguf](https://huggingface.co/second-state/Qwen3-14B-GGUF/blob/main/Qwen3-14B-Q5_K_S.gguf) | Q5_K_S | 5 | 10.3 GB| large, low quality loss - recommended |
-| [Qwen3-14B-Q6_K.gguf](https://huggingface.co/second-state/Qwen3-14B-GGUF/blob/main/Qwen3-14B-Q6_K.gguf)     | Q6_K   | 6 | 12.1 GB| very large, extremely low quality loss |
-| [Qwen3-14B-Q8_0.gguf](https://huggingface.co/second-state/Qwen3-14B-GGUF/blob/main/Qwen3-14B-Q8_0.gguf)     | Q8_0   | 8 | 15.7 GB| very large, extremely low quality loss - not recommended |
-| [Qwen3-14B-f16.gguf](https://huggingface.co/second-state/Qwen3-14B-GGUF/blob/main/Qwen3-14B-f16.gguf)       | f16    | 16 | 29.5 GB| |
 *Quantized with llama.cpp b5097*

+---
+base_model: Qwen/Qwen3-14B
+license: apache-2.0
+license_link: https://huggingface.co/Qwen/Qwen3-14B/blob/main/LICENSE
+model_creator: Qwen
+model_name: Qwen3-14B
+quantized_by: Second State Inc.
+language:
+- en
+pipeline_tag: text-generation
+tags:
+- chat
+library_name: transformers
+---
+<!-- 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 -->
+# Qwen3-14B-GGUF
+## Original Model
+[Qwen/Qwen3-14B](https://huggingface.co/Qwen/Qwen3-14B)
+## Run with LlamaEdge
+- LlamaEdge version:
+  - Thinking: [v0.17.0](https://github.com/LlamaEdge/LlamaEdge/releases/tag/0.17.0) and above
+  - No Thinking: [v0.18.2](https://github.com/LlamaEdge/LlamaEdge/releases/tag/0.18.2)
+- Prompt template
+  - Prompt type: `chatml` (for thinking)
+  - Prompt string
+    ```text
+    <|im_start|>system
+    {system_message}<|im_end|>
+    <|im_start|>user
+    {prompt}<|im_end|>
+    <|im_start|>assistant
+    ```
+  - Prompt type: `qwen3-no-think` (for no thinking)
+  - Prompt string
+    ```text
+    <|im_start|>system
+    {system_message}<|im_end|>
+    <|im_start|>user
+    {user_message_1}<|im_end|>
+    <|im_start|>assistant
+    {assistant_message_1}<|im_end|>
+    <|im_start|>user
+    {user_message_2}<|im_end|>
+    <|im_start|>assistant
+    <think>
+    </think>
+    ```
+- Context size: `128000`
+- Run as LlamaEdge service
+  ```bash
+  wasmedge --dir .:. --nn-preload default:GGML:AUTO:Qwen3-14B-Q5_K_M.gguf \
+    llama-api-server.wasm \
+    --model-name Qwen3-14B \
+    --prompt-template chatml \
+    --ctx-size 128000
+  ```
+- Run as LlamaEdge command app
+  ```bash
+  wasmedge --dir .:. --nn-preload default:GGML:AUTO:Qwen3-14B-Q5_K_M.gguf \
+    llama-chat.wasm \
+    --prompt-template chatml \
+    --ctx-size 128000
+  ```
+## Quantized GGUF Models
+| Name | Quant method | Bits | Size | Use case |
+| ---- | ---- | ---- | ---- | ----- |
+| [Qwen3-14B-Q2_K.gguf](https://huggingface.co/second-state/Qwen3-14B-GGUF/blob/main/Qwen3-14B-Q2_K.gguf)     | Q2_K   | 2 | 5.75 GB| smallest, significant quality loss - not recommended for most purposes |
+| [Qwen3-14B-Q3_K_L.gguf](https://huggingface.co/second-state/Qwen3-14B-GGUF/blob/main/Qwen3-14B-Q3_K_L.gguf) | Q3_K_L | 3 | 7.90 GB| small, substantial quality loss |
+| [Qwen3-14B-Q3_K_M.gguf](https://huggingface.co/second-state/Qwen3-14B-GGUF/blob/main/Qwen3-14B-Q3_K_M.gguf) | Q3_K_M | 3 | 7.32 GB| very small, high quality loss |
+| [Qwen3-14B-Q3_K_S.gguf](https://huggingface.co/second-state/Qwen3-14B-GGUF/blob/main/Qwen3-14B-Q3_K_S.gguf) | Q3_K_S | 3 | 6.66 GB| very small, high quality loss |
+| [Qwen3-14B-Q4_0.gguf](https://huggingface.co/second-state/Qwen3-14B-GGUF/blob/main/Qwen3-14B-Q4_0.gguf)     | Q4_0   | 4 | 8.52 GB| legacy; small, very high quality loss - prefer using Q3_K_M |
+| [Qwen3-14B-Q4_K_M.gguf](https://huggingface.co/second-state/Qwen3-14B-GGUF/blob/main/Qwen3-14B-Q4_K_M.gguf) | Q4_K_M | 4 | 9.00 GB| medium, balanced quality - recommended |
+| [Qwen3-14B-Q4_K_S.gguf](https://huggingface.co/second-state/Qwen3-14B-GGUF/blob/main/Qwen3-14B-Q4_K_S.gguf) | Q4_K_S | 4 | 8.57 GB| small, greater quality loss |
+| [Qwen3-14B-Q5_0.gguf](https://huggingface.co/second-state/Qwen3-14B-GGUF/blob/main/Qwen3-14B-Q5_0.gguf)     | Q5_0   | 5 | 10.3 GB| legacy; medium, balanced quality - prefer using Q4_K_M |
+| [Qwen3-14B-Q5_K_M.gguf](https://huggingface.co/second-state/Qwen3-14B-GGUF/blob/main/Qwen3-14B-Q5_K_M.gguf) | Q5_K_M | 5 | 10.5 GB| large, very low quality loss - recommended |
+| [Qwen3-14B-Q5_K_S.gguf](https://huggingface.co/second-state/Qwen3-14B-GGUF/blob/main/Qwen3-14B-Q5_K_S.gguf) | Q5_K_S | 5 | 10.3 GB| large, low quality loss - recommended |
+| [Qwen3-14B-Q6_K.gguf](https://huggingface.co/second-state/Qwen3-14B-GGUF/blob/main/Qwen3-14B-Q6_K.gguf)     | Q6_K   | 6 | 12.1 GB| very large, extremely low quality loss |
+| [Qwen3-14B-Q8_0.gguf](https://huggingface.co/second-state/Qwen3-14B-GGUF/blob/main/Qwen3-14B-Q8_0.gguf)     | Q8_0   | 8 | 15.7 GB| very large, extremely low quality loss - not recommended |
+| [Qwen3-14B-f16.gguf](https://huggingface.co/second-state/Qwen3-14B-GGUF/blob/main/Qwen3-14B-f16.gguf)       | f16    | 16 | 29.5 GB| |
 *Quantized with llama.cpp b5097*