Skip to main content
"""
Batch Agent-as-Judge Evaluation
===============================

Demonstrates evaluating multiple cases in one run.
"""

from agno.db.sqlite import SqliteDb
from agno.eval.agent_as_judge import AgentAsJudgeEval

# ---------------------------------------------------------------------------
# Create Database
# ---------------------------------------------------------------------------
db = SqliteDb(db_file="tmp/agent_as_judge_batch.db")

# ---------------------------------------------------------------------------
# Create Evaluation
# ---------------------------------------------------------------------------
evaluation = AgentAsJudgeEval(
    name="Customer Service Quality",
    criteria="Response should be empathetic, professional, and helpful",
    scoring_strategy="binary",
    db=db,
)

# ---------------------------------------------------------------------------
# Run Evaluation
# ---------------------------------------------------------------------------
if __name__ == "__main__":
    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_summary=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)}")

    print("Database Results:")
    eval_runs = db.get_eval_runs()
    print(f"Total evaluations stored: {len(eval_runs)}")
    if eval_runs:
        latest = eval_runs[-1]
        print(f"Eval ID: {latest.run_id}")
        print(f"Cases evaluated: {len(result.results)}")

​
Run the Example

# Clone and setup repo
git clone https://github.com/agno-agi/agno.git
cd agno/cookbook/09_evals/agent_as_judge

# Create and activate virtual environment
./scripts/demo_setup.sh
source .venvs/demo/bin/activate

python agent_as_judge_batch.py