Update app.py

app.py

@@ -4,6 +4,7 @@ from fastapi.responses import StreamingResponse, JSONResponse
 from pydantic import BaseModel
 from typing import List, Optional
 
+import asyncio
 import json
 import time
 import uuid
@@ -34,7 +35,7 @@ API_KEY = "sk-your-secret-key"
 
 app = FastAPI(
     title="Universal AI Gateway",
-    version="5.1.0"
+    version="5.2.0"
 )
 
 
@@ -106,12 +107,12 @@ async def root():
     return {
         "status": "online",
         "service": "Universal AI Gateway",
-        "version": "5.1.0"
+        "version": "5.2.0"
     }
 
 
 # =====================================================
-# GET MODELS
+# MODELS
 # =====================================================
 
 @app.get("/v1/models")
@@ -145,20 +146,34 @@ async def get_models():
 
 
 # =====================================================
-# SAFE COMPLETION
-# لا asyncio
-# لا to_thread
-# لا wait_for
+# NORMAL COMPLETION
+# =====================================================
+
+def normal_completion(model, messages):
+
+    client = Client()
+
+    response = client.chat.completions.create(
+        model=model,
+        messages=messages,
+        stream=False
+    )
+
+    return response
+
+
+# =====================================================
+# STREAM COMPLETION
 # =====================================================
 
-def safe_completion(model, messages, stream=False):
+def stream_completion(model, messages):
 
     client = Client()
 
     response = client.chat.completions.create(
         model=model,
         messages=messages,
-        stream=stream
+        stream=True
     )
 
     return response
@@ -200,10 +215,10 @@ async def chat_completions(
 
             try:
 
-                response = safe_completion(
-                    model=body.model,
-                    messages=messages,
-                    stream=True
+                response = await asyncio.to_thread(
+                    stream_completion,
+                    body.model,
+                    messages
                 )
 
                 has_content = False
@@ -337,10 +352,10 @@ async def chat_completions(
 
     try:
 
-        response = safe_completion(
-            model=body.model,
-            messages=messages,
-            stream=False
+        response = await asyncio.to_thread(
+            normal_completion,
+            body.model,
+            messages
         )
 
         assistant_message = ""