Upload app.py

app.py

@@ -1,23 +1,75 @@
 import os
+import re
 import gradio as gr
 import requests
 import inspect
 import pandas as pd
 
+from smolagents import (
+    CodeAgent,
+    InferenceClientModel,
+    DuckDuckGoSearchTool,
+    PythonInterpreterTool,
+)
+
 # (Keep Constants as is)
 # --- Constants ---
 DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
 
 # --- Basic Agent Definition ---
 # ----- THIS IS WERE YOU CAN BUILD WHAT YOU WANT ------
+
+# Free model available on HF's serverless Inference API.
+MODEL_ID = "Qwen/Qwen2.5-Coder-32B-Instruct"
+
+SYSTEM_PROMPT = """You are a general-purpose research and reasoning agent being
+graded on an exact-match benchmark. For every question you must:
+
+1. Think step by step and use your tools (web search, code execution) as
+   needed to find the correct answer.
+2. Once you are confident, respond with ONLY the final answer - no
+   explanation, no "The answer is", no restated question, no extra
+   punctuation or units unless the question explicitly asks for them.
+3. If the answer is a number, give just the number.
+4. If the answer is a short string, match the wording/capitalization implied
+   by the question as closely as possible.
+5. Never wrap your final answer in quotes or markdown formatting.
+"""
+
+
+def _clean_answer(raw: str) -> str:
+    """Strip common wrapper text so answers match exact-match grading."""
+    text = raw.strip()
+    text = re.sub(r"(?i)^\s*(final answer|answer)\s*[:\-]\s*", "", text)
+    if len(text) >= 2 and text[0] == text[-1] and text[0] in ("'", '"'):
+        text = text[1:-1]
+    return text.strip()
+
+
 class BasicAgent:
     def __init__(self):
         print("BasicAgent initialized.")
+        model = InferenceClientModel(
+            model_id=MODEL_ID,
+            token=os.environ.get("HF_TOKEN"),
+        )
+        self.agent = CodeAgent(
+            tools=[DuckDuckGoSearchTool(), PythonInterpreterTool()],
+            model=model,
+            add_base_tools=False,
+            max_steps=8,
+        )
+
     def __call__(self, question: str) -> str:
         print(f"Agent received question (first 50 chars): {question[:50]}...")
-        fixed_answer = "This is a default answer."
-        print(f"Agent returning fixed answer: {fixed_answer}")
-        return fixed_answer
+        try:
+            result = self.agent.run(SYSTEM_PROMPT + "\n\nQuestion:\n" + question)
+            answer = _clean_answer(str(result))
+        except Exception as e:
+            print(f"Agent error: {e}")
+            answer = ""
+        print(f"Agent returning answer: {answer}")
+        return answer
 
 def run_and_submit_all( profile: gr.OAuthProfile | None):
     """
@@ -146,11 +198,9 @@ with gr.Blocks() as demo:
     gr.Markdown(
         """
         **Instructions:**
-
         1.  Please clone this space, then modify the code to define your agent's logic, the tools, the necessary packages, etc ...
         2.  Log in to your Hugging Face account using the button below. This uses your HF username for submission.
         3.  Click 'Run Evaluation & Submit All Answers' to fetch questions, run your agent, submit answers, and see the score.
-
         ---
         **Disclaimers:**
         Once clicking on the "submit button, it can take quite some time ( this is the time for the agent to go through all the questions).
@@ -193,4 +243,4 @@ if __name__ == "__main__":
     print("-"*(60 + len(" App Starting ")) + "\n")
 
     print("Launching Gradio Interface for Basic Agent Evaluation...")
-    demo.launch(debug=True, share=False)
+    demo.launch(debug=True, share=False)