""" Enhanced Multi-LLM Agent Evaluation Runner with Agno Integration""" import os import gradio as gr import requests import pandas as pd from langchain_core.messages import HumanMessage # Import the enhanced classes from veryfinal.py in the same directory try: from veryfinal import ( build_graph, UnifiedAgnoEnhancedSystem, AgnoEnhancedAgentSystem, AgnoEnhancedModelManager ) VERYFINAL_AVAILABLE = True except ImportError as e: print(f"Error importing from veryfinal.py: {e}") VERYFINAL_AVAILABLE = False # --- Constants --- DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space" # --- Enhanced Agent Definition --- class EnhancedMultiLLMAgent: """A multi-provider Agno agent with NVIDIA + open-source model integration.""" def __init__(self): print("Enhanced Multi-LLM Agent with Agno Integration initialized.") if not VERYFINAL_AVAILABLE: print("Error: veryfinal.py not properly imported") self.system = None self.graph = None return try: # Use the unified Agno enhanced system self.system = UnifiedAgnoEnhancedSystem() self.graph = self.system.graph # Display system information if self.system.agno_system: info = self.system.get_system_info() print(f"System initialized with {info.get('total_models', 0)} models") if info.get('nvidia_available'): print("✅ NVIDIA NIM models available") print(f"Active agents: {info.get('active_agents', [])}") print("Enhanced Agno Multi-LLM System built successfully.") except Exception as e: print(f"Error building enhanced system: {e}") self.graph = None self.system = None def __call__(self, question: str) -> str: print(f"Agent received question: {question[:100]}...") if self.system is None: return "Error: Agent not properly initialized" try: # Use the enhanced system's process_query method answer = self.system.process_query(question) # Additional validation if not answer or answer == question or len(answer.strip()) == 0: return "Information not available" # Clean up the answer answer = answer.strip() # Ensure proper formatting for evaluation if "FINAL ANSWER:" in answer: answer = answer.split("FINAL ANSWER:")[-1].strip() return answer except Exception as e: error_msg = f"Error: {str(e)}" print(error_msg) return error_msg def run_and_submit_all(profile: gr.OAuthProfile | None): """Fetch questions, run enhanced Agno agent, and submit answers.""" space_id = os.getenv("SPACE_ID") if profile: username = f"{profile.username}" print(f"User logged in: {username}") else: print("User not logged in.") return "Please Login to Hugging Face with the button.", None api_url = DEFAULT_API_URL questions_url = f"{api_url}/questions" submit_url = f"{api_url}/submit" # 1. Instantiate Enhanced Agent try: agent = EnhancedMultiLLMAgent() if agent.system is None: return "Error: Failed to initialize enhanced agent properly", None except Exception as e: print(f"Error instantiating agent: {e}") return f"Error initializing agent: {e}", None agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main" if space_id else "No space ID available" print(f"Agent code URL: {agent_code}") # 2. Fetch Questions print(f"Fetching questions from: {questions_url}") try: response = requests.get(questions_url, timeout=15) response.raise_for_status() questions_data = response.json() if not questions_data: print("Fetched questions list is empty.") return "Fetched questions list is empty or invalid format.", None print(f"Fetched {len(questions_data)} questions.") except Exception as e: print(f"Error fetching questions: {e}") return f"Error fetching questions: {e}", None # 3. Run Enhanced Agno Agent results_log = [] answers_payload = [] print(f"Running Enhanced Agno Multi-LLM agent on {len(questions_data)} questions...") for i, item in enumerate(questions_data): task_id = item.get("task_id") question_text = item.get("question") if not task_id or question_text is None: print(f"Skipping item with missing task_id or question: {item}") continue print(f"Processing question {i+1}/{len(questions_data)}: {task_id}") try: submitted_answer = agent(question_text) # Additional validation to prevent question repetition if submitted_answer == question_text or submitted_answer.startswith(question_text): submitted_answer = "Information not available" answers_payload.append({"task_id": task_id, "submitted_answer": submitted_answer}) results_log.append({ "Task ID": task_id, "Question": question_text[:100] + "..." if len(question_text) > 100 else question_text, "Submitted Answer": submitted_answer[:200] + "..." if len(submitted_answer) > 200 else submitted_answer }) except Exception as e: error_msg = f"AGENT ERROR: {e}" print(f"Error running agent on task {task_id}: {e}") answers_payload.append({"task_id": task_id, "submitted_answer": error_msg}) results_log.append({ "Task ID": task_id, "Question": question_text[:100] + "..." if len(question_text) > 100 else question_text, "Submitted Answer": error_msg }) 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) # 4. Prepare Submission submission_data = {"username": username.strip(), "agent_code": agent_code, "answers": answers_payload} status_update = f"Enhanced Agno Multi-LLM Agent finished. Submitting {len(answers_payload)} answers for user '{username}'..." print(status_update) # 5. Submit print(f"Submitting {len(answers_payload)} answers to: {submit_url}") try: response = requests.post(submit_url, json=submission_data, timeout=60) response.raise_for_status() result_data = response.json() final_status = ( f"Submission Successful!\n" f"User: {result_data.get('username')}\n" f"Overall Score: {result_data.get('score', 'N/A')}% " f"({result_data.get('correct_count', '?')}/{result_data.get('total_attempted', '?')} correct)\n" f"Message: {result_data.get('message', 'No message received.')}" ) print("Submission successful.") results_df = pd.DataFrame(results_log) return final_status, results_df except Exception as e: status_message = f"Submission Failed: {e}" print(status_message) results_df = pd.DataFrame(results_log) return status_message, results_df # --- Build Gradio Interface --- with gr.Blocks() as demo: gr.Markdown("# Enhanced Multi-LLM Agent with Agno + NVIDIA Integration") gr.Markdown( """ **Instructions:** 1. Log in to your Hugging Face account using the button below. 2. Click 'Run Evaluation & Submit All Answers' to fetch questions, run your agent, submit answers, and see the score. **Enhanced Agent Features:** - **NVIDIA NIM Models**: Enterprise-grade optimized models for maximum accuracy - **Open-Source Models**: Groq, Ollama, Together AI, Anyscale, Hugging Face - **Specialized Agents**: Enterprise research, advanced math, coding, fast response - **Intelligent Routing**: Automatically selects best model/agent for each task - **Advanced Tools**: DuckDuckGo search, Wikipedia, calculator, reasoning tools - **Agno Framework**: Professional agent framework with memory and tool integration **Available Model Providers:** - **NVIDIA NIM**: meta/llama3-70b-instruct, meta/codellama-70b-instruct, etc. - **Groq (Free)**: llama3-70b-8192, llama3-8b-8192, mixtral-8x7b-32768 - **Ollama (Local)**: llama3, mistral, phi3, codellama, gemma, qwen - **Together AI**: Meta-Llama models, Mistral, Qwen - **Anyscale**: Enterprise hosting for open-source models - **Hugging Face**: Direct model access **Routing Examples:** - Enterprise: "Enterprise analysis of quantum computing" → NVIDIA NIM - Math: "Calculate 25 × 17" → Advanced Math Agent - Code: "Write Python factorial function" → Advanced Coding Agent - Research: "Find Mercedes Sosa discography" → Enterprise Research Agent - Quick: "Capital of France?" → Fast Response Agent **Setup Requirements:** - NVIDIA_API_KEY for enterprise models (optional) - GROQ_API_KEY for free tier models - Other API keys optional for additional providers """ ) gr.LoginButton() run_button = gr.Button("Run Evaluation & Submit All Answers", variant="primary") status_output = gr.Textbox(label="Run Status / Submission Result", lines=5, interactive=False) results_table = gr.DataFrame(label="Questions and Agent Answers", wrap=True) run_button.click( fn=run_and_submit_all, outputs=[status_output, results_table] ) if __name__ == "__main__": print("\n" + "-"*30 + " Enhanced Agno Multi-LLM Agent Starting " + "-"*30) # Display system status if VERYFINAL_AVAILABLE: try: test_system = UnifiedAgnoEnhancedSystem() info = test_system.get_system_info() print(f"✅ System ready with {info.get('total_models', 0)} models") print(f"📊 Model breakdown: {len(info.get('model_breakdown', {}).get('nvidia_models', []))} NVIDIA, " f"{len(info.get('model_breakdown', {}).get('groq_models', []))} Groq, " f"{len(info.get('model_breakdown', {}).get('ollama_models', []))} Ollama") except Exception as e: print(f"⚠️ System initialization warning: {e}") else: print("❌ veryfinal.py not properly imported") demo.launch(debug=True, share=False)