Minikube (mini-langstream)
To create a LangStream cluster locally, it's recommended to use minikube - setting 4 CPUs is highly recommended. mini-langstream
comes in help for installing and managing your local cluster.
Install mini-langstream
mini-langstream requires the following commands to be already installed:
Docker
Minikube
Helm
Kubectl
LangStream CLI
MacOS:
brew install LangStream/langstream/mini-langstream
Unix:
curl -Ls "https://raw.githubusercontent.com/LangStream/langstream/main/mini-langstream/get-mini-langstream.sh" | bash
Start the cluster
mini-langstream will do all the setup for you, in particular:
start
minikube
in a dedicated contextdeploy LangStream components using
helm
run a stateful Kafka broker as docker container
run a stateful s3-compatible storage (MinIO) as docker container
forward the control plane and API Gateways ports locally
create a dedicated LangStream CLI profile to interact with the cluster
wrap all the common k8s tools to inspect the cluster (
mini-langstream kubectl
,mini-langstream helm
,mini-langstream k9s
)
Start or ensure the cluster is running:
mini-langstream start
Try to use the CLI:
mini-langstream cli apps list
Deploy an application:
export OPENAI_API_KEY=<your-openai-api-key>
mini-langstream cli apps deploy my-app -app https://github.com/LangStream/langstream/tree/main/examples/applications/openai-completions -s https://github.com/LangStream/langstream/blob/main/examples/secrets/secrets.yaml
To delete all the storage and stop the cluster:
mini-langstream delete
Your first application
Here are a few ways to get started building LangStream applications:
Build a sample application to quickly create an OpenAI query with LangStream.
Set up your development environment and learn how to build the needed manifests
Install the VSCode extension and use the provided starter applications & agent snippets
Last updated