Upload app.py

app.py

@@ -3,10 +3,11 @@ import os
 import re
 from contextlib import asynccontextmanager
 
+
 from fastapi import FastAPI, HTTPException, Request
 from pydantic import BaseModel
 
-MODEL_NAME = "Qwen/Qwen3.5-9B"
+MODEL_NAME = "Qwen/Qwen3.5-9B:together"
 API_KEY = os.getenv("API_KEY")
 HF_TOKEN = os.getenv("HF_TOKEN")
 
@@ -25,7 +26,6 @@ async def lifespan(app: FastAPI):
         print("HF_TOKEN found — using HF Inference API")
         from huggingface_hub import InferenceClient
         inference_client = InferenceClient(
-            provider="together",
             api_key=HF_TOKEN,
         )
         print("Inference client ready.")
@@ -92,13 +92,13 @@ def run_hf_api(texts: list[str]) -> str:
         {"role": "system", "content": SYSTEM_PROMPT},
         {"role": "user", "content": build_prompt(texts)},
     ]
-    response = inference_client.chat_completion(
+    response = inference_client.chat.completions.create(
         model=MODEL_NAME,
         messages=messages,
         max_tokens=256,
         temperature=0.1,
     )
-    return response.choices[0].message.content
+    return response.choices[0].message.content or ""
 
 
 def run_local(texts: list[str]) -> str: