ci: add auto deployment for tabby live demo (TabbyML#2034)

.github/workflows/deploy-demo.yml

@@ -0,0 +1,38 @@
+name: Deploy to demo.tabbyml.com
+
+on:
+  workflow_dispatch:
+  release:
+    types: [ published ]
+
+jobs:
+  deploy:
+    name: Deploy
+    runs-on: ubuntu-latest
+    env:
+      MODAL_TOKEN_ID: ${{ secrets.MODAL_TOKEN_ID }}
+      MODAL_TOKEN_SECRET: ${{ secrets.MODAL_TOKEN_SECRET }}
+
+    steps:
+      - name: Checkout Repository
+        uses: actions/checkout@v4
+
+      - name: Install Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: "3.10"
+
+      - name: Install Modal
+        run: |
+          python -m pip install --upgrade pip
+          pip install modal
+
+      - name: Write image version
+        run: |
+          TABBY_IMAGE=tabbyml/tabby:${GITHUB_REF_NAME#v}
+          echo image $TABBY_IMAGE
+          echo "TABBY_IMAGE=$TABBY_IMAGE" >> $GITHUB_ENV
+
+      - name: Deploy job
+        run: |
+          cd experimental/demo modal deploy app.py

experimental/demo/app.py

@@ -0,0 +1,64 @@
+import os
+from modal import Image, Stub, gpu, asgi_app, Volume
+
+IMAGE_NAME = os.environ.get("TABBY_IMAGE", "tabbyml/tabby")
+
+image = (
+    Image.from_registry(
+        IMAGE_NAME,
+        add_python="3.11",
+    )
+    .dockerfile_commands("ENTRYPOINT []")
+    .pip_install("asgi-proxy-lib")
+)
+
+stub = Stub("tabby-demo-server", image=image)
+volume = Volume.from_name("tabby-demo-server-volume", create_if_missing=True)
+
+@stub.function(
+    concurrency_limit=1,
+    allow_concurrent_inputs=100,
+    container_idle_timeout=600*2,
+    timeout=600,
+    volumes = {"/data": volume},
+    _allow_background_volume_commits=True
+)
+@asgi_app()
+def entry():
+    import socket
+    import subprocess
+    import time
+    import os
+    from asgi_proxy import asgi_proxy
+
+    env = os.environ.copy()
+    env["TABBY_DISABLE_USAGE_COLLECTION"] = "1"
+    env["TABBY_WEBSERVER_DEMO_MODE"] = "1"
+    launcher = subprocess.Popen(
+        [
+            "/opt/tabby/bin/tabby-cpu",
+            "serve",
+            "--port",
+            "8000",
+        ],
+        env=env
+    )
+
+    # Poll until webserver at 127.0.0.1:8000 accepts connections before running inputs.
+    def tabby_ready():
+        try:
+            socket.create_connection(("127.0.0.1", 8000), timeout=1).close()
+            return True
+        except (socket.timeout, ConnectionRefusedError):
+            # Check if launcher webserving process has exited.
+            # If so, a connection can never be made.
+            retcode = launcher.poll()
+            if retcode is not None:
+                raise RuntimeError(f"launcher exited unexpectedly with code {retcode}")
+            return False
+
+    while not tabby_ready():
+        time.sleep(1.0)
+
+    print("Tabby server ready!")
+    return asgi_proxy("http://localhost:8000")