Adjusted rate limiting

app.py

@@ -36,7 +36,20 @@ app.state.limiter = limiter
 async def custom_rate_limit_handler(request: Request, exc: RateLimitExceeded):
     client_ip = get_remote_address(request)
     logger.warning(f"Rate limit exceeded for IP {client_ip} on endpoint {request.url.path}")
-    return await _rate_limit_exceeded_handler(request, exc)
+    
+    # Return a proper JSON response for rate limiting
+    return Response(
+        content=json.dumps({
+            "error": "rate_limit_exceeded",
+            "message": "Too many requests. Please wait a moment before trying again.",
+            "retry_after": 60  # seconds
+        }),
+        status_code=429,
+        headers={
+            "Content-Type": "application/json",
+            "Retry-After": "60"
+        }
+    )
 
 app.add_exception_handler(RateLimitExceeded, custom_rate_limit_handler)
 
@@ -346,14 +359,14 @@ def get_ram_usage():
     }
 
 @app.post("/api/predict")
-@limiter.limit("5/minute")  # Allow 10 chat requests per minute per IP
+@limiter.limit("15/minute")  # Allow 15 chat requests per minute per IP
 async def predict(request: Request, query_request: QueryRequest):
     query = query_request.query
     history = query_request.history
     return StreamingResponse(stream_response(query, history), media_type="text/event-stream")
 
 @app.post("/api/tts")
-@limiter.limit("5/minute")  # Allow 5 TTS requests per minute per IP (more restrictive as TTS is more expensive)
+@limiter.limit("10/minute")  # Allow 10 TTS requests per minute per IP
 async def text_to_speech(request: Request, tts_request: TTSRequest):
     """Convert text to speech using ElevenLabs API"""
     if not elevenlabs_client: