SGLang: install nightly wheel from the CUDA 13 index
#6
by Jiminator - opened
README.md
CHANGED
|
@@ -146,10 +146,8 @@ Laguna M.1 can be served with SGLang using its OpenAI-compatible server, includi
|
|
| 146 |
> Laguna support was added to SGLang in [sgl-project/sglang#24204](https://github.com/sgl-project/sglang/pull/24204). The integration is shared with [Laguna XS.2](https://huggingface.co/poolside/Laguna-XS.2) and is currently available on SGLang main.
|
| 147 |
|
| 148 |
```shell
|
| 149 |
-
# Laguna M.1 support is currently on SGLang main
|
| 150 |
-
|
| 151 |
-
cd sglang
|
| 152 |
-
pip install -e "python[all]"
|
| 153 |
|
| 154 |
sglang serve \
|
| 155 |
--model-path poolside/Laguna-M.1 \
|
|
|
|
| 146 |
> Laguna support was added to SGLang in [sgl-project/sglang#24204](https://github.com/sgl-project/sglang/pull/24204). The integration is shared with [Laguna XS.2](https://huggingface.co/poolside/Laguna-XS.2) and is currently available on SGLang main.
|
| 147 |
|
| 148 |
```shell
|
| 149 |
+
# Laguna M.1 support is currently on SGLang main; install the latest nightly wheel
|
| 150 |
+
pip install sglang --pre --extra-index-url https://sgl-project.github.io/whl/cu130/
|
|
|
|
|
|
|
| 151 |
|
| 152 |
sglang serve \
|
| 153 |
--model-path poolside/Laguna-M.1 \
|