Sambanova Basic
Run Llama 3.3 70B on SambaCloud with sync, streaming, and async calls.
SambaNova removed Meta-Llama-3.1-8B-Instruct from SambaCloud on April 14, 2026. Apply the replacement below before running. SambaStack deployments have separate availability.
"""
Sambanova Basic
===============
Cookbook example for `sambanova/basic.py`.
"""
from agno.agent import Agent, RunOutput # noqa
from agno.models.sambanova import Sambanova
import asyncio
# ---------------------------------------------------------------------------
# Create Agent
# ---------------------------------------------------------------------------
agent = Agent(model=Sambanova(id="Meta-Llama-3.1-8B-Instruct"), markdown=True)
# Get the response in a variable
# run: RunOutput = agent.run("Share a 2 sentence horror story")
# print(run.content)
# Print the response in the terminal
# ---------------------------------------------------------------------------
# Run Agent
# ---------------------------------------------------------------------------
if __name__ == "__main__":
# --- Sync ---
agent.print_response("Share a 2 sentence horror story")
# --- Sync + Streaming ---
agent.print_response("Share a 2 sentence horror story", stream=True)
# --- Async ---
asyncio.run(agent.aprint_response("Share a 2 sentence horror story"))
# --- Async + Streaming ---
asyncio.run(agent.aprint_response("Share a 2 sentence horror story", stream=True))Run the Example
Set up your virtual environment
uv venv --python 3.12
source .venv/bin/activateInstall dependencies
uv pip install -U agno openaiExport your SambaNova API key
export SAMBANOVA_API_KEY="your_sambanova_api_key_here"Select the current SambaCloud model
Replace Meta-Llama-3.1-8B-Instruct with Meta-Llama-3.3-70B-Instruct in each saved model constructor. Confirm your SambaCloud key has access to this recommended replacement.
Run the example
Save the code above as basic.py, then run:
python basic.pyFull source: cookbook/90_models/sambanova/basic.py