Update utils.py

utils.py

@@ -40,10 +40,13 @@ Question:
 Answer (with explanation):
 """
 
-    client = InferenceClient(token=os.environ.get("HUGGINGFACE_API_KEY"))
+    # ✅ Explicitly specify provider (avoids StopIteration)
+    client = InferenceClient(
+        provider="hf-inference",
+        token=os.environ.get("HUGGINGFACE_API_KEY")
+    )
 
     try:
-        # ✅ Works for huggingface-hub >=0.33
         response = client.text_generation(
             prompt,
             model="google/gemma-2b-it",
@@ -51,14 +54,12 @@ Answer (with explanation):
             temperature=0.7,
         )
     except TypeError:
-        # ✅ Fallback for older versions that use 'inputs' & 'parameters'
         response = client.text_generation(
             model="google/gemma-2b-it",
             inputs=prompt,
             parameters={"max_new_tokens": 1024, "temperature": 0.7},
         )
 
-    # Extract text safely
     if isinstance(response, str):
         answer = response
     elif isinstance(response, dict) and "generated_text" in response:
@@ -68,4 +69,4 @@ Answer (with explanation):
     else:
         answer = str(response)
 
-    return answer
+    return answer