Update app.py

app.py

@@ -10,10 +10,19 @@ from typing import Optional
 logging.basicConfig(level=logging.DEBUG)
 logger = logging.getLogger(__name__)
 
-app = FastAPI()
+app = FastAPI(
+    title="Orion AI API",
+    description="API for streaming AI responses with model selection and publisher via URL",
+    version="1.0.0"
+)
 
-# Define valid models (replace with actual models supported by your AI server)
-VALID_MODELS = ["default-model", "another-model"]  # Update this list
+# Define valid models (replace with actual models supported by https://models.github.ai/inference)
+VALID_MODELS = [
+    "deepseek/DeepSeek-V3-0324",  # Added based on your request
+    "gpt-3.5-turbo",              # Common model (placeholder)
+    "llama-3",                    # Common model (placeholder)
+    "mistral-7b"                  # Common model (placeholder)
+]
 
 class GenerateRequest(BaseModel):
     prompt: str
@@ -46,7 +55,7 @@ async def generate_ai_response(prompt: str, model: str, publisher: Optional[str]
     client = AsyncOpenAI(base_url=endpoint, api_key=token)
 
     try:
-        # Include publisher in the request payload (modify as needed based on AI server requirements)
+        # Include publisher in the request payload
         stream = await client.chat.completions.create(
             messages=[
                 {"role": "system", "content": "You are a helpful assistant named Orion, created by Abdullah Ali"},
@@ -70,7 +79,7 @@ async def generate_ai_response(prompt: str, model: str, publisher: Optional[str]
 
 @app.post("/generate", summary="Generate AI response", response_description="Streaming AI response")
 async def generate_response(
-    model: str = Query("default-model", description="The AI model to use"),
+    model: str = Query("deepseek/DeepSeek-V3-0324", description="The AI model to use"),
     prompt: Optional[str] = Query(None, description="The input text prompt for the AI"),
     publisher: Optional[str] = Query(None, description="Publisher identifier (optional, defaults to DEFAULT_PUBLISHER env var)"),
     request: Optional[GenerateRequest] = None
@@ -78,7 +87,7 @@ async def generate_response(
     """
     Generate a streaming AI response based on the provided prompt, model, and publisher.
     
-    - **model**: The AI model to use (e.g., default-model)
+    - **model**: The AI model to use (e.g., deepseek/DeepSeek-V3-0324)
     - **prompt**: The input text prompt for the AI (query param or body)
     - **publisher**: The publisher identifier (optional, defaults to DEFAULT_PUBLISHER env var)
     """
@@ -102,5 +111,12 @@ async def generate_response(
         media_type="text/event-stream"
     )
 
+@app.get("/models", summary="List available models")
+async def list_models():
+    """
+    List all available models supported by the AI server.
+    """
+    return {"models": VALID_MODELS}
+
 def get_app():
     return app