Streaming Basic Agent

Code

1from typing import Iterator
2from kern.agent import Agent, RunOutput
3from kern.models.ibm import WatsonX
4
5agent = Agent(model=WatsonX(id="ibm/granite-20b-code-instruct"), markdown=True)
6
7# Get the response in a variable
8# run_response: Iterator[RunOutputEvent] = agent.run("Share a 2 sentence horror story", stream=True)
9# for chunk in run_response:
10#     print(chunk.content)
11
12# Print the response in the terminal
13agent.print_response("Share a 2 sentence horror story", stream=True)

Usage

Set up your virtual environment

1uv venv --python 3.12
2source .venv/bin/activate

1uv venv --python 3.12
2.venv\Scripts\activate

Set your API key

1export IBM_WATSONX_API_KEY=xxx
2export IBM_WATSONX_PROJECT_ID=xxx

Install dependencies

1uv pip install -U ibm-watsonx-ai kern-ai

Run Agent

1python cookbook/11_models/ibm/watsonx/basic_stream.py

This example shows how to use streaming with IBM WatsonX. Setting stream=True when calling print_response() or run() enables token-by-token streaming, which can provide a more interactive user experience.