Streaming

Stream tokens in real-time as the model generates them. Instead of waiting for the full response, your app receives each token as it's produced.

Stream to console

Pass stream=True to print tokens to the console as they arrive:

stream.py

from veska import Agent

agent = Agent(
    name="writer",
    system_prompt="You are a creative writer.",
    model="claude-sonnet-4-6",
)

result = agent.run("Write a poem about the ocean", stream=True)

Stream to a custom function

Pass a function to stream to handle each token yourself:

stream_custom.py

from veska import Agent

agent = Agent(
    name="writer",
    system_prompt="You are a creative writer.",
    model="claude-sonnet-4-6",
)

# Send tokens to your UI, file, or anywhere
def on_token(text):
    my_ui.append(text)

result = agent.run("Write a poem about the ocean", stream=on_token)

# result still contains the full response
print(result.output)

Stream options

Value	Behavior
`None / False`	No streaming. Waits for full response.
`True`	Prints tokens to console automatically.
`function`	Calls your function with each token.

Provider support

Both Claude and OpenAI models support streaming. The behavior is the same regardless of which model you use.