Batch Agent as Judge

Evaluate multiple input/output pairs in a single batch

This example demonstrates batch evaluation where multiple cases are evaluated together.

1. Add the following code to your Python file

   agent_as_judge_batch.py

   from agno.db.sqlite import SqliteDb
   from agno.eval.agent_as_judge import AgentAsJudgeEval
   
   # Setup database to persist eval results
   db = SqliteDb(db_file="tmp/agent_as_judge_batch.db")
   
   evaluation = AgentAsJudgeEval(
       name="Customer Service Quality",
       criteria="Response should be empathetic, professional, and helpful",
       scoring_strategy="binary",  # PASS/FAIL for each case
       db=db,
   )
   
   result = evaluation.run(
       cases=[
           {
               "input": "My order is delayed and I'm very upset!",
               "output": "I sincerely apologize for the delay. I understand how frustrating this must be. Let me check your order status right away and see how we can make this right for you.",
           },
           {
               "input": "Can you help me with a refund?",
               "output": "Of course! I'd be happy to help with your refund. Could you please provide your order number so I can process this quickly for you?",
           },
           {
               "input": "Your product is terrible!",
               "output": "I'm sorry to hear you're disappointed. Your feedback is valuable to us. Could you share more details about what went wrong so we can improve?",
           },
       ],
       print_results=True,
   )
   
   print(f"Pass rate: {result.pass_rate:.1f}%")
   print(f"Passed: {sum(1 for r in result.results if r.passed)}/{len(result.results)}")
   
   
2. Set up your virtual environment

   Mac

   uv venv --python 3.12
   source .venv/bin/activate
   
3. Install dependencies

   uv pip install -U agno openai
   
4. Export your OpenAI API key

   Mac/Linux

     export OPENAI_API_KEY="your_openai_api_key_here"
   
5. Run the example

   python agent_as_judge_batch.py