dfdafda

proxy_server.py

@@ -1412,27 +1412,11 @@ async def completion(
         raise HTTPException(status_code=status, detail=error_msg)
 
 
-@router.post("/chat/completions", dependencies=[Depends(user_api_key_auth)])
-async def completion(request: Request):
-    key = request.headers.get("Authorization").replace("Bearer ", "")  # type: ignore
-    data = await request.json()
-    print(f"received request data: {data}")
-    data["user_key"] = key
-    data["budget_manager"] = budget_manager
-    data["master_key"] = master_key
-    set_env_variables(data)
-    # handle how users send streaming
-    if 'stream' in data:
-        if type(data['stream']) == str: # if users send stream as str convert to bool
-            # convert to bool
-            if data['stream'].lower() == "true":
-                data['stream'] = True # convert to boolean
-    
-    response = llm.completion(**data)
-    if 'stream' in data and data['stream'] == True: # use generate_responses to stream responses
-            return StreamingResponse(data_generator(response), media_type='text/event-stream')
-    return response
-    
+@router.post(
+    "/v1/chat/completions",
+    dependencies=[Depends(user_api_key_auth)],
+    tags=["chat/completions"],
+)
 @router.post(
     "/chat/completions",
     dependencies=[Depends(user_api_key_auth)],
@@ -2534,7 +2518,7 @@ async def config_yaml_endpoint(config_info: ConfigYAML):
 
     print(chat_completion.choices[0].message.content)  
 
-@router.get("/assistant/chat", tags=["openinterpreter"])
+@router.get("/openinterpreter/chat", tags=["openinterpreter"])
 async def config_yaml_endpoint(config_info: ConfigYAML):
     client = Groq(
         api_key=os.environ.get("GROQ_API_KEY"),
@@ -2551,7 +2535,6 @@ async def config_yaml_endpoint(config_info: ConfigYAML):
     )
 
     print(chat_completion.choices[0].message.content) 
-    return chat_completion.choices[0].message.content
 
 
 @router.get("/test", tags=["health"])