config.yaml, iterate with truss watch, and push to Baseten. It also runs the Chains, Training, and Loops workflows. Install with uvx truss or pip install truss.
The Baseten CLI Beta manages your workspace: organizations, API keys, secrets, and the deployment lifecycle, from promoting deployments to terminating replicas. It also calls models and the Baseten API. Every command supports --output json and --jq filtering, built for scripts and CI. Install with Homebrew.
Both CLIs are open source and authenticate against the same Baseten backend.
Which one do I use?
| Task | CLI |
|---|---|
Authoring model.py or config.yaml and iterating with truss watch | Truss |
| Deploying a model from your editor | Truss (truss push) |
| Deploying a model from a CI workflow that parses output | Baseten CLI (baseten model push) |
| Authoring a Chain, Training job, or Loop | Truss (truss chains, truss train, truss loops) |
| Creating or deleting an org API key | Baseten CLI (baseten org api-key) |
| Promoting a development deployment to production | Baseten CLI (baseten model deployment promote) |
| Querying a model from a shell pipeline | Baseten CLI (baseten model predict --jq ...) |
| Managing org secrets | Baseten CLI (baseten org secret) |
| Terminating a deployment replica | Baseten CLI (baseten model deployment replica) |
Overlapping commands
A few commands exist in both CLIs:| Action | Truss | Baseten CLI | Pick based on |
|---|---|---|---|
| Authenticate | truss login | baseten auth login | Whichever CLI you already have installed. Both hit the same backend. |
| Deploy a model | truss push | baseten model push | truss push while iterating locally (supports --watch). baseten model push from CI (structured output, scriptable errors). |
| Tail logs | truss model-logs | baseten model deployment logs | Match the workflow you’re already in. |
Use both together
baseten truss forwards arguments to the truss binary on your PATH, so baseten truss push works as a synonym for truss push.
Next steps
Author a model
Use Truss to package and deploy your first model.
Manage your workspace
Use the Baseten CLI for orgs, deployments, and automation.
Build a Chain
Create multi-step inference pipelines with
truss chains.Wire CI/CD
Use
baseten model push in GitHub Actions and other pipelines.