Fixed the streaming issue

app.py

@@ -223,7 +223,6 @@ async def vercelXaigenerate(json_data: Dict[str, Any]):
         'user-agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/135.0.0.0 Safari/537.36'
     }
 
-    # Convert messages to XAI format
     messages = conver_to_xai_schema(json_data["messages"])
     
     request_data = {
@@ -248,7 +247,12 @@ async def vercelXaigenerate(json_data: Dict[str, Any]):
                     async for line in request_ctx.aiter_lines():
                         if line:
                             if line.startswith('0:'):
-                                text = line[2:]
+                                # Clean up the text and properly escape JSON characters
+                                text = line[2:].strip()
+                                if text.startswith('"') and text.endswith('"'):
+                                    text = text[1:-1]
+                                text = text.replace('\\n', '\n').replace('\\', '')
+                                
                                 response = {
                                     "id": chunk_id,
                                     "object": "chat.completion.chunk",

logs.json

@@ -13,5 +13,150 @@
     "timestamp": "2025-04-12T11:51:54.228166",
     "endpoint": "/chat",
     "query": "vercelXaigenerate"
+  },
+  {
+    "timestamp": "2025-04-12T11:59:54.231283",
+    "endpoint": "/chat",
+    "query": "vercelXaigenerate"
+  },
+  {
+    "timestamp": "2025-04-12T12:00:40.844595",
+    "endpoint": "/chat",
+    "query": "vercelXaigenerate"
+  },
+  {
+    "timestamp": "2025-04-12T12:00:45.093082",
+    "endpoint": "/chat",
+    "query": "generate"
+  },
+  {
+    "timestamp": "2025-04-12T12:00:53.603673",
+    "endpoint": "/chat",
+    "query": "vercelXaigenerate"
+  },
+  {
+    "timestamp": "2025-04-12T12:01:08.563064",
+    "endpoint": "/chat",
+    "query": "vercelXaigenerate"
+  },
+  {
+    "timestamp": "2025-04-12T12:01:11.525946",
+    "endpoint": "/chat",
+    "query": "generate"
+  },
+  {
+    "timestamp": "2025-04-12T12:01:14.278740",
+    "endpoint": "/chat",
+    "query": "vercelXaigenerate"
+  },
+  {
+    "timestamp": "2025-04-12T12:01:16.915702",
+    "endpoint": "/chat",
+    "query": "vercelXaigenerate"
+  },
+  {
+    "timestamp": "2025-04-12T12:01:30.033848",
+    "endpoint": "/chat",
+    "query": "groqgenerate"
+  },
+  {
+    "timestamp": "2025-04-12T12:02:15.117096",
+    "endpoint": "/chat",
+    "query": "vercelXaigenerate"
+  },
+  {
+    "timestamp": "2025-04-12T12:02:23.649030",
+    "endpoint": "/chat",
+    "query": "generate"
+  },
+  {
+    "timestamp": "2025-04-12T12:02:28.933558",
+    "endpoint": "/chat",
+    "query": "groqgenerate"
+  },
+  {
+    "timestamp": "2025-04-12T12:02:38.160605",
+    "endpoint": "/chat",
+    "query": "generate"
+  },
+  {
+    "timestamp": "2025-04-12T12:02:52.076861",
+    "endpoint": "/chat",
+    "query": "vercelXaigenerate"
+  },
+  {
+    "timestamp": "2025-04-12T12:04:02.320142",
+    "endpoint": "/chat",
+    "query": "generate"
+  },
+  {
+    "timestamp": "2025-04-12T12:04:05.385644",
+    "endpoint": "/chat",
+    "query": "vercelXaigenerate"
+  },
+  {
+    "timestamp": "2025-04-12T12:04:20.976615",
+    "endpoint": "/chat",
+    "query": "generate"
+  },
+  {
+    "timestamp": "2025-04-12T12:04:24.590270",
+    "endpoint": "/chat",
+    "query": "groqgenerate"
+  },
+  {
+    "timestamp": "2025-04-12T12:04:34.063593",
+    "endpoint": "/chat",
+    "query": "groqgenerate"
+  },
+  {
+    "timestamp": "2025-04-12T12:04:38.292885",
+    "endpoint": "/chat",
+    "query": "generate"
+  },
+  {
+    "timestamp": "2025-04-12T12:04:40.796568",
+    "endpoint": "/chat",
+    "query": "vercelXaigenerate"
+  },
+  {
+    "timestamp": "2025-04-12T12:05:13.292850",
+    "endpoint": "/chat",
+    "query": "groqgenerate"
+  },
+  {
+    "timestamp": "2025-04-12T12:05:23.222112",
+    "endpoint": "/chat",
+    "query": "groqgenerate"
+  },
+  {
+    "timestamp": "2025-04-12T12:05:28.215528",
+    "endpoint": "/chat",
+    "query": "groqgenerate"
+  },
+  {
+    "timestamp": "2025-04-12T12:05:32.617464",
+    "endpoint": "/chat",
+    "query": "vercelXaigenerate"
+  },
+  {
+    "timestamp": "2025-04-12T12:06:42.995442",
+    "endpoint": "/chat",
+    "query": "generate"
+  },
+  {
+    "timestamp": "2025-04-12T12:06:44.829603",
+    "endpoint": "/chat",
+    "query": "groqgenerate"
+  },
+  {
+    "timestamp": "2025-04-12T12:06:52.601184",
+    "endpoint": "/chat",
+    "query": "groqgenerate"
+  },
+  {
+    "timestamp": "2025-04-12T12:06:56.874556",
+    "endpoint": "/chat",
+    "query": "vercelXaigenerate"
   }
 ]

test.py

@@ -0,0 +1,32 @@
+import requests
+import json
+
+url = "http://127.0.0.1:8000/chat"
+
+payload = {
+    "model": "your-model-name",
+    "message": "Create a json object of top 10 anime .json response only!",
+    "messages": []
+}
+
+headers = {
+    "Content-Type": "application/json"
+}
+
+# Send streaming POST request
+with requests.post(url, data=json.dumps(payload), headers=headers, stream=True) as response:
+    if response.status_code == 200:
+        for line in response.iter_lines(decode_unicode=True):
+            if line and line.startswith('data: '):
+                try:
+                    # Remove 'data: ' prefix and parse JSON
+                    json_data = json.loads(line[6:])
+                    # Extract text from choices if available
+                    if json_data.get('choices') and len(json_data['choices']) > 0:
+                        text = json_data['choices'][0].get('text', '')
+                        if text:
+                            print(text, end='')
+                except json.JSONDecodeError:
+                    continue
+    else:
+        print("Error:", response.status_code, response.text)