Update app.py

app.py

@@ -1,12 +1,13 @@
 from fastapi import FastAPI, HTTPException
 from fastapi.middleware.cors import CORSMiddleware
-from fastapi.responses import StreamingResponse, JSONResponse
+from fastapi.responses import JSONResponse
 import os
 from dotenv import load_dotenv
 import requests
 from typing import Dict, Any, List
 from pydantic import BaseModel
 import time
+import json
 
 load_dotenv()
 
@@ -43,9 +44,21 @@ def clean_assistant_response(text: str) -> str:
         text = text[:-3].strip()
     return text
 
+class CustomJSONResponse(JSONResponse):
+    media_type = "application/json; charset=utf-8"
+
+    def render(self, content: Any) -> bytes:
+        return json.dumps(
+            content,
+            ensure_ascii=False,
+            allow_nan=False,
+            indent=None,
+            separators=(',', ':')
+        ).encode('utf-8')
+
 @app.get("/")
 async def root():
-    return {"status": "FastFlowWrapper is running"}
+    return CustomJSONResponse({"status": "FastFlowWrapper is running"})
 
 @app.get("/v1/models")
 async def get_models():
@@ -61,15 +74,15 @@ async def get_models():
             models.append({
                 "id": chatflow.get("id"),
                 "object": "model",
-                "created": int(time.time()),  # Текущий timestamp
+                "created": int(time.time()),
                 "owned_by": "flowise",
                 "permission": [],
                 "root": "flowise",
                 "parent": None,
-                "system_fingerprint": "phi4-r1"  # Добавляем system_fingerprint
+                "system_fingerprint": "phi4-r1"
             })
         
-        return {"object": "list", "data": models}
+        return CustomJSONResponse({"object": "list", "data": models})
     except requests.RequestException as e:
         raise HTTPException(status_code=500, detail=str(e))
 
@@ -92,10 +105,11 @@ async def create_chat_completion(request: ChatCompletionRequest):
         # Засекаем время начала запроса
         start_time = time.time()
         
-        # Отправляем запрос к Flowise
+        # Отправляем запрос к Flowise с таймаутом
         response = requests.post(
             f"{FLOWISE_API_BASE_URL}/prediction/{FLOWISE_CHATFLOW_ID}",
-            json=flowise_request
+            json=flowise_request,
+            timeout=10  # Уменьшаем таймаут до 10 секунд
         )
         response.raise_for_status()
         
@@ -106,15 +120,15 @@ async def create_chat_completion(request: ChatCompletionRequest):
         # Подсчитываем токены ответа
         completion_tokens = count_tokens(assistant_response)
         
-        return {
+        return CustomJSONResponse({
             "id": "chatcmpl-" + os.urandom(12).hex(),
             "object": "chat.completion",
-            "created": int(start_time),  # Используем время начала запроса
-            "model": "phi4-r1",  # Используем тот же model_id что и в прямом API
+            "created": int(start_time),
+            "model": "phi4-r1",
             "choices": [
                 {
                     "index": 0,
-                    "logprobs": None,  # Добавляем поле logprobs
+                    "logprobs": None,
                     "finish_reason": "stop",
                     "message": {
                         "role": "assistant",
@@ -127,8 +141,8 @@ async def create_chat_completion(request: ChatCompletionRequest):
                 "completion_tokens": completion_tokens,
                 "total_tokens": prompt_tokens + completion_tokens
             },
-            "stats": {},  # Добавляем пустой объект stats
-            "system_fingerprint": "phi4-r1"  # Добавляем system_fingerprint
-        }
+            "stats": {},
+            "system_fingerprint": "phi4-r1"
+        })
     except requests.RequestException as e:
         raise HTTPException(status_code=500, detail=str(e))