Update app.py

app.py

@@ -20,8 +20,6 @@ from langchain_community.chat_models import ChatHuggingFace
 # --- Constants ---
 DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
 
-hf_token = os.getenv("HF_TOKEN")
-
 # --- Basic Agent Definition ---
 # ----- THIS IS WERE YOU CAN BUILD WHAT YOU WANT ------
 class BasicAgent:
@@ -35,13 +33,27 @@ class BasicAgent:
 
 class ZephyrPipelineModel:
     def __init__(self):
-        self.pipe = pipeline("text-generation", model="HuggingFaceH4/zephyr-7b-beta")
+        hf_token = os.getenv("HF_TOKEN")
+        if not hf_token:
+            raise ValueError("HF_TOKEN environment variable not set.")
+            
+        self.client = InferenceClient(
+            model="HuggingFaceH4/zephyr-7b-beta",
+            token=hf_token,
+        )
 
     def __call__(self, prompt: str) -> str:
-        # Zephyr wants a list of messages
         messages = [{"role": "user", "content": prompt}]
-        result = self.pipe(messages, max_new_tokens=512)
-        return result[0]["generated_text"]
+        try:
+            completion = self.client.chat.completions.create(
+                model="HuggingFaceH4/zephyr-7b-beta",
+                messages=messages,
+                max_tokens=512,
+                temperature=0.7,
+            )
+            return completion.choices[0].message.content
+        except Exception as e:
+            return f"❌ Inference failed: {str(e)}"
 
 class LangGraphAgent:
     def __init__(self):