Update to use LiteLLM proxy server configuration

api.py

@@ -1,42 +0,0 @@
-from fastapi import FastAPI
-from fastapi.staticfiles import StaticFiles
-from fastapi.middleware.cors import CORSMiddleware
-from fastapi.openapi.docs import get_swagger_ui_html
-from litellm.proxy.proxy_server import app as proxy_app
-
-app = FastAPI(
-    title="LiteLLM API",
-    version="1.0.0",
-)
-
-# Add CORS middleware
-app.add_middleware(
-    CORSMiddleware,
-    allow_origins=["*"],
-    allow_credentials=True,
-    allow_methods=["*"],
-    allow_headers=["*"],
-)
-
-# Mount static files
-app.mount("/static", StaticFiles(directory="static"), name="static")
-
-# Mount the LiteLLM Proxy server
-app.mount("/proxy", proxy_app)
-
-@app.get("/", include_in_schema=False)
-async def custom_swagger_ui_html():
-    return get_swagger_ui_html(
-        openapi_url="/openapi.json",
-        title="LiteLLM API",
-        swagger_js_url="/static/swagger/swagger-ui-bundle.js",
-        swagger_css_url="/static/swagger/swagger-ui.css",
-    )
-
-@app.get("/health")
-def health_check():
-    return {"status": "healthy"}
-
-if __name__ == "__main__":
-    import uvicorn
-    uvicorn.run(app, host="0.0.0.0", port=7860)

app.py

@@ -1,11 +1,25 @@
 from fastapi import FastAPI
-from litellm.proxy.proxy_server import app as proxy_app
+from litellm.proxy.proxy_server import ProxyServer
+from litellm.proxy.config import ProxyConfig
 
-app = FastAPI()
+app = FastAPI(
+    title="LiteLLM Proxy",
+    description="LiteLLM OpenAI-compatible proxy",
+    version="1.0",
+    docs_url="/proxy/docs",         # Swagger UI
+    redoc_url="/proxy/redoc",       # Optional: ReDoc UI
+    openapi_url="/proxy/openapi.json"
+)
 
-# Mount the LiteLLM Proxy server
-app.mount("/proxy", proxy_app)
+# Load LiteLLM Proxy
+proxy_config = ProxyConfig()
+proxy_server = ProxyServer(config=proxy_config)
+proxy_server.add_routes(app)
 
 @app.get("/")
-def home():
-    return {"message": "LiteLLM Server is running! Visit /proxy/docs for the API documentation"}
+async def root():
+    return {"message": "LiteLLM is running. Visit /proxy/docs"}
+
+if __name__ == "__main__":
+    import uvicorn
+    uvicorn.run(app, host="0.0.0.0", port=7860)

config.yaml

@@ -3,13 +3,22 @@ model_list:
     litellm_params:
       model: gpt-3.5-turbo
 
+server_settings:
+  port: 7860
+  host: "0.0.0.0"
+  environment: "production"
+
 general_settings:
+  # Master key for admin access
   master_key: "sk-1234"
-  port: 8000  # Changed to 8080 for Hugging Face spaces
-  # Enable UI dashboard
+  # Enable UI features
   ui_features:
     analytics_dashboard: true
     model_config_management: true
     key_management: true
-  allow_origins: ["*"]  # Allow cross-origin requests
-  start_server: true
+  # Allow all origins for CORS
+  allow_origins: ["*"]
+  # Other configurations
+  save_responses: true
+  cache_responses: false
+  log_requests: true