update install commands

Dockerfile

@@ -123,6 +123,11 @@ RUN --mount=type=bind,from=build,src=/workspace/dist,target=/vllm-workspace/dist
     --mount=type=cache,target=/root/.cache/pip \
     pip install dist/*.whl --verbose
 
+# UPSTREAM SYNC: Install sparsity extras
+RUN --mount=type=bind,from=build,src=/workspace/dist,target=/vllm-workspace/dist \
+    --mount=type=cache,target=/root/.cache/pip \
+    pip install nm-magic-wand-nightly --extra-index-url https://pypi.neuralmagic.com/simple
+
 RUN --mount=type=bind,from=flash-attn-builder,src=/usr/src/flash-attention-v2,target=/usr/src/flash-attention-v2 \
     --mount=type=cache,target=/root/.cache/pip \
     pip install /usr/src/flash-attention-v2/*.whl --no-cache-dir
@@ -135,10 +140,6 @@ FROM vllm-base AS test
 
 ADD . /vllm-workspace/
 
-# UPSTREAM SYNC: Install sparsity extras
-RUN --mount=type=cache,target=/root/.cache/pip \
-    pip3 install nm-magic-wand-nightly
-
 # install development dependencies (for testing)
 RUN --mount=type=cache,target=/root/.cache/pip \
     pip install -r requirements-dev.txt

README.md

@@ -12,12 +12,12 @@ The [nm-vllm PyPi package](https://pypi.org/project/nm-vllm/) includes pre-compi
 
 Install it using pip:
 ```bash
-pip install nm-vllm
+pip install nm-vllm --extra-index-url https://pypi.neuralmagic.com/simple
 ```
 
 For utilizing weight-sparsity kernels, such as through `sparsity="sparse_w16a16"`, you can extend the installation with the `sparsity` extras:
 ```bash
-pip install nm-vllm[sparse]
+pip install nm-vllm[sparse] --extra-index-url https://pypi.neuralmagic.com/simple
 ```
 
 You can also build and install `nm-vllm` from source (this will take ~10 minutes):

examples-neuralmagic/deploy_compressed_huggingface_models/Deploy_Compressed_LLMs_from_Hugging_Face_with_nm_vllm.ipynb

@@ -5386,7 +5386,7 @@
         }
       ],
       "source": [
-        "!pip install nm-vllm[sparse]"
+        "!pip install nm-vllm[sparse] --extra-index-url https://pypi.neuralmagic.com/simple"
       ]
     },
     {
@@ -6219,4 +6219,4 @@
       }
     }
   ]
-}
+}

examples-neuralmagic/marlin_quantization_and_deploy/Performantly_Quantize_LLMs_to_4_bits_with_Marlin_and_nm_vllm.ipynb

@@ -1506,7 +1506,7 @@
         }
       ],
       "source": [
-        "!pip install nm-vllm"
+        "!pip install nm-vllm --extra-index-url https://pypi.neuralmagic.com/simple"
       ]
     },
     {

examples-neuralmagic/sparsegpt_compress_and_deploy/Apply_SparseGPT_to_LLMs_and_deploy_with_nm_vllm.ipynb

@@ -7899,7 +7899,7 @@
     {
       "cell_type": "code",
       "source": [
-        "!pip install nm-vllm[sparse]"
+        "!pip install nm-vllm[sparse] --extra-index-url https://pypi.neuralmagic.com/simple"
       ],
       "metadata": {
         "id": "Sz2Cs4BtPa7_"
@@ -8062,4 +8062,4 @@
       "outputs": []
     }
   ]
-}
+}