combined_metrics.py
"""
Combined Metrics
=============================
When an agent uses multiple background features, each model's
calls are tracked under separate detail keys:
- "model" for the agent's own calls
- "reasoning_model" for reasoning manager calls
- "compression_model" for compression manager calls
- "output_model" for output model calls
- "memory_model" for memory manager calls
- "culture_model" for culture manager calls
- "session_summary_model" for session summary calls
- "eval_model" for evaluation hook calls
This example shows all detail keys and session-level metrics.
"""
from typing import List
from agno.agent import Agent
from agno.compression.manager import CompressionManager
from agno.culture.manager import CultureManager
from agno.db.postgres import PostgresDb
from agno.eval.agent_as_judge import AgentAsJudgeEval
from agno.memory.manager import MemoryManager
from agno.models.openai import OpenAIChat
from agno.session.summary import SessionSummaryManager
from agno.tools.yfinance import YFinanceTools
from pydantic import BaseModel, Field
from rich.pretty import pprint
class StockSummary(BaseModel):
ticker: str = Field(..., description="Stock ticker symbol")
summary: str = Field(..., description="Brief summary of the stock")
key_metrics: List[str] = Field(..., description="Key financial metrics")
# ---------------------------------------------------------------------------
# Create Agent
# ---------------------------------------------------------------------------
db = PostgresDb(db_url="postgresql+psycopg://ai:ai@localhost:5532/ai")
eval_hook = AgentAsJudgeEval(
name="Quality Check",
model=OpenAIChat(id="gpt-4o-mini"),
criteria="Response should be helpful and accurate",
scoring_strategy="binary",
)
agent = Agent(
model=OpenAIChat(id="gpt-4o-mini"),
tools=[YFinanceTools(enable_stock_price=True, enable_company_info=True)],
reasoning_model=OpenAIChat(id="gpt-4o-mini"),
reasoning=True,
compression_manager=CompressionManager(
model=OpenAIChat(id="gpt-4o-mini"),
compress_tool_results_limit=1,
),
output_model=OpenAIChat(id="gpt-4o-mini"),
output_schema=StockSummary,
structured_outputs=True,
memory_manager=MemoryManager(model=OpenAIChat(id="gpt-4o-mini"), db=db),
update_memory_on_run=True,
culture_manager=CultureManager(model=OpenAIChat(id="gpt-4o-mini"), db=db),
update_cultural_knowledge=True,
session_summary_manager=SessionSummaryManager(model=OpenAIChat(id="gpt-4o-mini")),
enable_session_summaries=True,
post_hooks=[eval_hook],
db=db,
session_id="combined-metrics-demo",
)
# ---------------------------------------------------------------------------
# Run Agent
# ---------------------------------------------------------------------------
if __name__ == "__main__":
run_response = agent.run(
"Get the stock price and company info for NVDA and summarize it."
)
print("=" * 50)
print("RUN METRICS")
print("=" * 50)
pprint(run_response.metrics)
print("=" * 50)
print("MODEL DETAILS")
print("=" * 50)
if run_response.metrics and run_response.metrics.details:
for model_type, model_metrics_list in run_response.metrics.details.items():
print(f"\n{model_type}:")
for model_metric in model_metrics_list:
pprint(model_metric)
print("=" * 50)
print("SESSION METRICS")
print("=" * 50)
session_metrics = agent.get_session_metrics()
if session_metrics:
pprint(session_metrics)
Run the Example
1
Set up your virtual environment
uv venv --python 3.12
source .venv/bin/activate
uv venv --python 3.12
.venv\Scripts\activate
2
Install dependencies
uv pip install -U agno "psycopg[binary]" openai sqlalchemy yfinance
3
Export your OpenAI API key
export OPENAI_API_KEY="your_openai_api_key_here"
$Env:OPENAI_API_KEY="your_openai_api_key_here"
4
Run PgVector
docker run -d \
-e POSTGRES_DB=ai \
-e POSTGRES_USER=ai \
-e POSTGRES_PASSWORD=ai \
-e PGDATA=/var/lib/postgresql \
-v pgvolume:/var/lib/postgresql \
-p 5532:5432 \
--name pgvector \
agnohq/pgvector:18
docker run -d `
-e POSTGRES_DB=ai `
-e POSTGRES_USER=ai `
-e POSTGRES_PASSWORD=ai `
-e PGDATA=/var/lib/postgresql `
-v pgvolume:/var/lib/postgresql `
-p 5532:5432 `
--name pgvector `
agnohq/pgvector:18
5
Run the example
Save the code above as
combined_metrics.py, then run:python combined_metrics.py