run and submit locally

agent.py

@@ -22,31 +22,35 @@ from tools import python_wikipedia_tools, smolagents_code_tools, smolagents_web_
 #            custom_role_conversions=None,
 #        )
 
-model = OpenAIServerModel(
-    model_id="gemini-2.0-flash",
-    api_base="https://generativelanguage.googleapis.com/v1beta/openai/",
-    api_key=os.getenv("GEMINI_API_KEY"),
-)
-
-# to run locally
 #model = OpenAIServerModel(
-#    model_id="gemma3:27b",
-#    api_base="http://localhost:11434/v1",
+#    model_id="gemini-2.0-flash",
+#    api_base="https://generativelanguage.googleapis.com/v1beta/openai/",
+#    api_key=os.getenv("GEMINI_API_KEY"),
 #)
 
+# to run locally
+model_gemma = LiteLLMModel(
+    model_id="ollama/gemma3:27b",
+    api_base="http://localhost:11434",
+)
+model_devstral = LiteLLMModel(
+    model_id="ollama/devstral:latest",
+    api_base="http://localhost:11434",
+)
+
 
-web_search_agent = CodeAgent(
+web_search_agent = ToolCallingAgent(
     tools=smolagents_web_tools,
-    model=model,
+    model=model_gemma,
     max_steps=10,
-    additional_authorized_imports=["wikipedia", "markdownify", "requests"],
+    #additional_authorized_imports=["wikipedia", "markdownify", "requests"],
     name="web_search_agent",
-    description="A code agent which can perform web searches. It has available tools including GoogleSearchTool, VisitWebpageTool and WikipediaSearchTool",
+    description="A tool calling agent which can perform web searches. It has available tools including GoogleSearchTool, VisitWebpageTool and WikipediaSearchTool",
 )
 
 code_agent = CodeAgent(
     tools=smolagents_code_tools + python_wikipedia_tools,
-    model=model,
+    model=model_gemma,
     max_steps=10,
     additional_authorized_imports=["time", "numpy", "pandas", "wikipedia", "xlrd", "markdownify", "requests"],
     name="code_agent",
@@ -55,10 +59,10 @@ code_agent = CodeAgent(
 
 manager_agent = CodeAgent(
     tools=[],
-    model=model,
+    model=model_gemma,
     managed_agents=[web_search_agent, code_agent],
     max_steps=10,
     verbosity_level=1,
     name="manager_agent",
-    description="Manages the web_search_agent and code_agent.",
+    description="Manages the web_search_agent and code_agent. Can perform web searches with the ToolCallingAgent called web_search_agent. Can use the CodeAgent called code_agent to run pythin code.",
 )

app.py

@@ -4,7 +4,7 @@ import requests
 import inspect
 import pandas as pd
 from agent import manager_agent
-from time import sleep
+#from time import sleep
 from prompts import prompt
 # --- Constants ---
 DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
@@ -75,7 +75,7 @@ def run_and_submit_all( profile: gr.OAuthProfile | None):
     results_log = []
     answers_payload = []
     print(f"Running agent on {len(questions_data)} questions...")
-    for item in questions_data[:1]: # First question for testing
+    for item in questions_data: # First question for testing
         task_id = item.get("task_id")
         question_text = item.get("question")
         if not task_id or question_text is None:
@@ -88,7 +88,7 @@ def run_and_submit_all( profile: gr.OAuthProfile | None):
         except Exception as e:
             print(f"Error running agent on task {task_id}: {e}")
             results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": f"AGENT ERROR: {e}"})
-        sleep(30) # 30 seconds delay between questions for Gemini
+        #sleep(30) # 30 seconds delay between questions for Gemini
     if not answers_payload:
         print("Agent did not produce any answers to submit.")
         return "Agent did not produce any answers to submit.", pd.DataFrame(results_log)
@@ -195,7 +195,7 @@ 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)
 
     #agent = BasicAgent()
     #question = "You are a general AI assistant. You will be asked a question. Report your thoughts, and finish your answer with the following template: FINAL ANSWER: [YOUR FINAL ANSWER]. YOUR FINAL ANSWER should be a number OR as few words as possible OR a comma separated list of numbers and/or strings. If you are asked for a number, don't use comma to write your number neither use units such as $ or percent sign unless specified otherwise. If you are asked for a string, don't use articles, neither abbreviations (e.g. for cities), and write the digits in plain text unless specified otherwise. If you are asked for a comma separated list, apply the above rules depending of whether the element to be put in the list is a number or a string. Think step by step and reason through the question. Then return your final answer.\n\nQuestion: How many studio albums were published by Mercedes Sosa between 2000 and 2009 (included)? You can use the latest 2022 version of english wikipedia."

prompts.py

@@ -1,8 +1,8 @@
 prompt = """You are a general AI assistant. You will be asked a question. 
-    Report your thoughts, and finish your answer with the following template: FINAL ANSWER: [YOUR FINAL ANSWER]. 
-    YOUR FINAL ANSWER should be a number OR as few words as possible OR a comma separated list of numbers and/or strings. 
+    Think step by step, and finish your answer with your FINAL ANSWER. 
+    Your FINAL ANSWER should be a number OR as few words as possible OR a comma separated list of numbers and/or strings. 
     If you are asked for a number, don't use comma to write your number neither use units such as $ or percent sign unless specified otherwise. 
     If you are asked for a string, don't use articles, neither abbreviations (e.g. for cities), and write the digits in plain text unless specified otherwise. 
     If you are asked for a comma separated list, apply the above rules depending of whether the element to be put in the list is a number or a string.
-    Think step by step and reason through the question. Then return your final answer.
+    Report only the value of your final answer, don't add any additional words.
 """

requirements.txt

@@ -1,9 +1,10 @@
 gradio
 requests
-smolagents[openai,transformers,litellm]
+smolagents[openai,transformers,litellm]>=1.17.0
 duckduckgo_search
 wikipedia-api
 wikipedia
 xlrd
 transformers
-markdownify
+markdownify
+aiohttp>=3.8.0