Skip to main content

​
Overview

The show() function (aliased as show_dashboard) opens the ZenML dashboard in your default web browser. This provides quick access to the ZenML UI for viewing pipelines, artifacts, models, and more.

​
Signature

from zenml import show

def show(
    local: bool = False,
    ngrok_token: Optional[str] = None,
) -> None:
    """Open the ZenML dashboard in your browser.
    
    Args:
        local: Whether to show local server dashboard
        ngrok_token: Token for exposing dashboard via ngrok
    """

​
Parameters

​
local
bool
default:"False"
If True, shows the dashboard for the local ZenML server. If False, shows the dashboard for the currently active server (could be remote).
​
ngrok_token
str
default:"None"
An ngrok authentication token to expose the ZenML dashboard on a public domain. Primarily useful for accessing the dashboard in cloud notebooks like Google Colab.

​
Return Value

​
return
None
This function does not return a value. It opens the dashboard in your default web browser.

​
Usage Examples

​
Open Active Server Dashboard

from zenml import show

# Open the dashboard for the currently active server
show()

​
Open Local Server Dashboard

from zenml import show

# Explicitly open the local server dashboard
show(local=True)

​
Use in Jupyter Notebooks

from zenml import show

# After running a pipeline
training_pipeline()

# View results in the dashboard
show()

​
Expose Dashboard with ngrok

from zenml import show

# Useful in Google Colab or cloud notebooks
show(ngrok_token="your_ngrok_auth_token")

​
Quick Pipeline Debugging

from zenml import pipeline, step, show

@step
def train_model() -> str:
    return "model_v1"

@pipeline
def training_pipeline():
    model = train_model()

if __name__ == "__main__":
    # Run pipeline
    training_pipeline()
    
    # Open dashboard to view results
    show()

​
Dashboard Features

When you open the dashboard, you can:

View Pipeline Runs

See all pipeline executions with status, duration, and artifacts

Explore Artifacts

Browse artifacts with lineage, metadata, and visualizations

Monitor Models

Track model versions, stages, and performance metrics

Manage Stacks

Configure and switch between different infrastructure stacks

View DAGs

Visualize pipeline structure and step dependencies

Compare Runs

Compare metrics and parameters across multiple runs

​
Server Connection

The show() function requires an active connection to a ZenML server:
1

Check Connection

Verify you’re connected to a server:
zenml status
2

Connect to Server

If not connected, login to a server:
# Local server
zenml login --local

# Remote server
zenml login https://your-server.com
3

Open Dashboard

Now you can use the show() function:
from zenml import show
show()

​
Errors

This error occurs when ZenML is not connected to any server. Fix by running:
zenml login --local
  • Check your default browser settings
  • The dashboard URL is still printed to console - copy and paste manually
  • Verify the server is running: zenml status

​
CLI Alternative

You can also open the dashboard via CLI: 
# Show active server dashboard
zenml show

# Show local server dashboard
zenml show --local

get_pipeline_context

Access pipeline information programmatically

get_step_context

Access step information in code

CLI: zenml show

Command-line dashboard access

​
Notes

The show() function is aliased as show_dashboard in the ZenML imports. Both names work identically.
In cloud notebook environments (Colab, Databricks), use the ngrok_token parameter to expose the dashboard via a public URL.
The ngrok token should be kept secure. Don’t commit it to version control or share it publicly.

Build docs developers (and LLMs) love

Get started for freeTalk to us