here goes ... nightly's almost there

.github/actions/nm-set-env/action.yml

@@ -1,6 +1,9 @@
 name: set neuralmagic env
 description: 'sets environment variables for neuralmagic'
 inputs:
+  wf_category:
+    description: "categories: REMOTE, NIGHTLY, RELEASE"
+    required: true
   hf_token:
     description: 'Hugging Face home'
     required: true
@@ -14,19 +17,27 @@ runs:
   using: composite
   steps:
   - run: |
+      # setup.py defaults to making 'nightly' package with 'nightly' version
+      if [[ "${{inputs.wf_category}}" == "RELEASE" ]]; then
+        echo "NM_RELEASE_TYPE=${{inputs.wf_category}}" >> $GITHUB_ENV
+      fi
+      # CUDA
       echo "TORCH_CUDA_ARCH_LIST=7.0 7.5 8.0 8.6 8.9 9.0+PTX" >> $GITHUB_ENV
+      echo "PATH=/usr/local/apps/pyenv/plugins/pyenv-virtualenv/shims:/usr/local/apps/pyenv/shims:/usr/local/apps/pyenv/bin:/usr/local/apps/nvm/versions/node/v19.9.0/bin:/usr/local/apps/nvm/versions/node/v16.20.2/bin:/usr/local/cuda-12.1/bin:/usr/local/cuda-12.1/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/snap/bin:/home/${WHOAMI}/.local/bin:" >> $GITHUB_ENV
+      echo "LD_LIBRARY_PATH=/usr/local/cuda-12.1/lib64::/usr/local/cuda-12.1/lib64:" >> $GITHUB_ENV
+      # HF Cache
       echo "HF_TOKEN=${HF_TOKEN_SECRET}" >> $GITHUB_ENV
       echo "HF_HOME=/EFS/hf_home" >> $GITHUB_ENV
+      # build
       NUM_THREADS=$(./.github/scripts/determine-threading -G ${{ inputs.Gi_per_thread }})
       echo "MAX_JOBS=${NUM_THREADS}" >> $GITHUB_ENV
       echo "NVCC_THREADS=${{ inputs.nvcc_threads }}" >> $GITHUB_ENV
       echo "VLLM_INSTALL_PUNICA_KERNELS=1" >> $GITHUB_ENV
       echo "NCCL_IGNORE_DISABLED_P2P=1" >> $GITHUB_ENV
+      # pyenv
       echo "PYENV_ROOT=/usr/local/apps/pyenv" >> $GITHUB_ENV
+      # testmo
       echo "XDG_CONFIG_HOME=/usr/local/apps" >> $GITHUB_ENV
-      WHOAMI=$(whoami)
-      echo "PATH=/usr/local/apps/pyenv/plugins/pyenv-virtualenv/shims:/usr/local/apps/pyenv/shims:/usr/local/apps/pyenv/bin:/usr/local/apps/nvm/versions/node/v19.9.0/bin:/usr/local/apps/nvm/versions/node/v16.20.2/bin:/usr/local/cuda-12.1/bin:/usr/local/cuda-12.1/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/snap/bin:/home/${WHOAMI}/.local/bin:" >> $GITHUB_ENV
-      echo "LD_LIBRARY_PATH=/usr/local/cuda-12.1/lib64::/usr/local/cuda-12.1/lib64:" >> $GITHUB_ENV
       echo "PROJECT_ID=12" >> $GITHUB_ENV
     env:
         HF_TOKEN_SECRET: ${{ inputs.hf_token }}

.github/workflows/build-test.yml

@@ -116,6 +116,7 @@ jobs:
     BUILD:
         uses: ./.github/workflows/build.yml
         with:
+            wf_category: ${{ inputs.wf_category }}
             build_label: ${{ inputs.build_label }}
             timeout: ${{ inputs.build_timeout }}
             gitref: ${{ github.ref }}

.github/workflows/build.yml

@@ -91,6 +91,7 @@ jobs:
               id: setenv
               uses: ./.github/actions/nm-set-env/
               with:
+                wf_category: ${{ inputs.wf_category }}
                 hf_token: ${{ secrets.NM_HF_TOKEN }}
                 Gi_per_thread: ${{ inputs.Gi_per_thread }}
                 nvcc_threads: ${{ inputs.nvcc_threads }}
@@ -111,11 +112,12 @@ jobs:
                 testmo_token: ${{ secrets.TESTMO_TEST_TOKEN }}
                 source: 'build-test'
 
-            - name: rename for nightly
-              uses: ./.github/actions/nm-rename-for-nightly-whl/
-              if: contains(fromJSON('["NIGHTLY"]'), inputs.wf_category)
-              with:
-                package_name: "nm-vllm"
+            # skip this ... should no be handled by setup.py
+            # - name: rename for nightly
+            #   uses: ./.github/actions/nm-rename-for-nightly-whl/
+            #   if: contains(fromJSON('["NIGHTLY"]'), inputs.wf_category)
+            #   with:
+            #     package_name: "nm-vllm"
 
             - name: build
               id: build

setup.py

@@ -2,6 +2,7 @@
 # UPSTREAM SYNC: noqa is required for passing ruff.
 # This file has been modified by Neural Magic
 
+import datetime
 import importlib.util
 import io
 import logging
@@ -306,9 +307,28 @@ def find_version(filepath: str) -> str:
         raise RuntimeError("Unable to find version string.")
 
 
+# Neuralmagic packaging ENV's
+NM_RELEASE_TYPE = 'NM_RELEASE_TYPE'
+
+
+def get_nm_vllm_package_name() -> str:
+    nm_release_type = os.getenv(NM_RELEASE_TYPE)
+    package_name = None
+    if nm_release_type == 'RELEASE':
+        package_name = 'nm-vllm'
+    else:
+        package_name = 'nm-vllm-nightly'
+    return package_name
+
+
 def get_vllm_version() -> str:
     version = find_version(get_path("vllm", "__init__.py"))
 
+    nm_release_type = os.getenv(NM_RELEASE_TYPE)
+    if nm_release_type != 'RELEASE':
+        date = datetime.date.today().strftime("%Y%m%d")
+        version += f'.{date}'
+
     if _is_cuda():
         cuda_version = str(get_nvcc_cuda_version())
         if cuda_version != MAIN_CUDA_VERSION:
@@ -393,6 +413,9 @@ def _read_requirements(filename: str) -> List[str]:
 
 # UPSTREAM SYNC: needed for sparsity
 _sparsity_deps = ["nm-magic-wand-nightly"]
+nm_release_type = os.getenv(NM_RELEASE_TYPE)
+if nm_release_type == 'RELEASE':
+    _sparsity_deps = ["nm-magic-wand"]
 
 package_data = {
     "vllm": ["py.typed", "model_executor/layers/fused_moe/configs/*.json"]
@@ -402,7 +425,7 @@ def _read_requirements(filename: str) -> List[str]:
     package_data["vllm"].append("*.so")
 
 setup(
-    name="nm-vllm",
+    name=get_nm_vllm_package_name(),
     version=get_vllm_version(),
     author="vLLM Team, Neural Magic",
     author_email="support@neuralmagic.com",