Update app.py

app.py

@@ -6,8 +6,7 @@ app = FastAPI()
 
 pipe = pipeline(
     "text-generation",
-    model="microsoft/phi-3-mini-4k-instruct",
-    device_map="auto"
+    model="microsoft/phi-3-mini-4k-instruct"
 )
 
 class Req(BaseModel):
@@ -17,7 +16,7 @@ def build_prompt(user_msg: str) -> str:
     return (
         "<|system|>\n"
         "You are a helpful, clear, and concise assistant. "
-        "Give accurate and well-structured answers. Avoid unnecessary fluff."
+        "Give accurate and well-structured answers."
         "<|end|>\n"
         "<|user|>\n"
         f"{user_msg}"
@@ -35,12 +34,10 @@ def chat(req: Req):
         temperature=0.6,
         top_p=0.9,
         do_sample=True,
-        repetition_penalty=1.1,
-        eos_token_id=pipe.tokenizer.eos_token_id
+        repetition_penalty=1.1
     )
 
     text = out[0]["generated_text"]
-
     reply = text.split("<|assistant|>")[-1].strip()
 
     return {"reply": reply}