Knowledge

Answer recipe questions using a Token Factory model and a PgVector knowledge base.

Token Factory lists Qwen/Qwen3-30B-A3B as retired on November 3, 2025. Select an enabled replacement using the steps below; see the model retirement notice.

knowledge.py
"""Run `uv pip install ddgs sqlalchemy pgvector pypdf cerebras_cloud_sdk` to install dependencies."""

from agno.agent import Agent
from agno.knowledge.knowledge import Knowledge
from agno.models.nebius import Nebius
from agno.vectordb.pgvector import PgVector

# ---------------------------------------------------------------------------
# Create Agent
# ---------------------------------------------------------------------------

db_url = "postgresql+psycopg://ai:ai@localhost:5532/ai"

knowledge = Knowledge(
    vector_db=PgVector(table_name="recipes", db_url=db_url),
)
# Add content to the knowledge
knowledge.insert(url="https://agno-public.s3.amazonaws.com/recipes/ThaiRecipes.pdf")

agent = Agent(model=Nebius(id="Qwen/Qwen3-30B-A3B"), knowledge=knowledge)
agent.print_response("How to make Thai curry?", markdown=True)

# ---------------------------------------------------------------------------
# Run Agent
# ---------------------------------------------------------------------------

if __name__ == "__main__":
    pass

Run the Example

Set up your virtual environment

uv venv --python 3.12
source .venv/bin/activate

Install dependencies

uv pip install -U agno "psycopg[binary]" beautifulsoup4 openai pgvector pypdf sqlalchemy

Export your API keys

export NEBIUS_API_KEY="your_nebius_api_key_here"
export OPENAI_API_KEY="your_openai_api_key_here"

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

Select an enabled Token Factory model

Copy a model ID available to your account for https://api.tokenfactory.nebius.com/v1 from the Token Factory console. For knowledge and tool examples it must support function calling; for structured output it must support JSON Schema. Confirm the endpoint and model's access before running.

export NEBIUS_MODEL_ID="your-enabled-model-id"

Replace the placeholder with your copied model ID. Add from os import environ to the saved file, then set every model construction to Nebius(id=environ["NEBIUS_MODEL_ID"]). Keep any additional model options. The adapter's default ID does not establish which models your account can access. A dedicated deployment must instead use the model and endpoint supplied for that deployment, with an explicit base_url.

Run the example

Save the code above as knowledge.py, then run:

python knowledge.py

Full source: cookbook/90_models/nebius/knowledge.py