Adding docker-compose.yaml, starting to simplify

distributions/remote-vllm/run-with-safety.yaml

@@ -0,0 +1,68 @@
+version: '2'
+built_at: '2024-11-11T20:09:45.988375'
+image_name: remote-vllm
+docker_image: remote-vllm
+conda_env: null
+apis:
+- inference
+- memory
+- safety
+- agents
+- telemetry
+providers:
+  inference:
+  # serves main inference model
+  - provider_id: vllm-inference
+    provider_type: remote::vllm
+    config:
+      # NOTE: replace with "localhost" if you are running in "host" network mode
+      url: ${env.VLLM_URL}
+      max_tokens: ${env.MAX_TOKENS:4096}
+      api_token: fake
+  # serves safety llama_guard model
+  - provider_id: vllm-safety
+    provider_type: remote::vllm
+    config:
+      # NOTE: replace with "localhost" if you are running in "host" network mode
+      url: ${env.SAFETY_VLLM_URL}
+      max_tokens: ${env.MAX_TOKENS:4096}
+      api_token: fake
+  memory:
+  - provider_id: faiss-0
+    provider_type: inline::faiss
+    config:
+      kvstore:
+        namespace: null
+        type: sqlite
+        db_path: "${env.SQLITE_STORE_DIR:~/.llama/distributions/remote-vllm}/faiss_store.db"
+  safety:
+  - provider_id: llama-guard
+    provider_type: inline::llama-guard
+    config: {}
+  memory:
+  - provider_id: meta0
+    provider_type: inline::faiss
+    config: {}
+  agents:
+  - provider_id: meta0
+    provider_type: inline::meta-reference
+    config:
+      persistence_store:
+        namespace: null
+        type: sqlite
+        db_path: "${env.SQLITE_STORE_DIR:~/.llama/distributions/remote-vllm}/agents_store.db"
+  telemetry:
+  - provider_id: meta0
+    provider_type: inline::meta-reference
+    config: {}
+metadata_store:
+  namespace: null
+  type: sqlite
+  db_path: "${env.SQLITE_STORE_DIR:~/.llama/distributions/remote-vllm}/registry.db"
+models:
+  - model_id: ${env.INFERENCE_MODEL}
+    provider_id: vllm-inference
+  - model_id: ${env.SAFETY_MODEL}
+    provider_id: vllm-safety
+shields:
+  - shield_id: ${env.SAFETY_MODEL}