Update README.md
Browse files
README.md
CHANGED
@@ -83,22 +83,24 @@ print("thinking content:", thinking_content)
|
|
83 |
print("content:", content)
|
84 |
```
|
85 |
|
86 |
-
For deployment, you can use `
|
87 |
-
-
|
88 |
```shell
|
89 |
-
|
90 |
```
|
91 |
-
-
|
92 |
```shell
|
93 |
-
|
94 |
```
|
95 |
|
|
|
|
|
|
|
96 |
## Switching Between Thinking and Non-Thinking Mode
|
97 |
|
98 |
> [!TIP]
|
99 |
-
>
|
100 |
-
>
|
101 |
-
> Please refer to our documentation for [vLLM](https://qwen.readthedocs.io/en/latest/deployment/vllm.html#thinking-non-thinking-modes) and [SGLang](https://qwen.readthedocs.io/en/latest/deployment/sglang.html#thinking-non-thinking-modes) users.
|
102 |
|
103 |
### `enable_thinking=True`
|
104 |
|
|
|
83 |
print("content:", content)
|
84 |
```
|
85 |
|
86 |
+
For deployment, you can use `sglang>=0.4.6.post1` or `vllm>=0.8.4` or to create an OpenAI-compatible API endpoint:
|
87 |
+
- SGLang:
|
88 |
```shell
|
89 |
+
python -m sglang.launch_server --model-path Qwen/Qwen3-32B --reasoning-parser qwen3
|
90 |
```
|
91 |
+
- vLLM:
|
92 |
```shell
|
93 |
+
vllm serve Qwen/Qwen3-32B --enable-reasoning --reasoning-parser deepseek_r1
|
94 |
```
|
95 |
|
96 |
+
For local use, applications such as llama.cpp, Ollama, LMStudio, and MLX-LM have also supported Qwen3.
|
97 |
+
|
98 |
+
|
99 |
## Switching Between Thinking and Non-Thinking Mode
|
100 |
|
101 |
> [!TIP]
|
102 |
+
> The `enable_thinking` switch is also available in APIs created by SGLang and vLLM.
|
103 |
+
> Please refer to our documentation for [SGLang](https://qwen.readthedocs.io/en/latest/deployment/sglang.html#thinking-non-thinking-modes) and [vLLM](https://qwen.readthedocs.io/en/latest/deployment/vllm.html#thinking-non-thinking-modes) users.
|
|
|
104 |
|
105 |
### `enable_thinking=True`
|
106 |
|