Deploying to Heroku

Heroku is a popular platform as a service based on managed container system. It provides an easy solution to quickly build, run and scale applications. BentoML works great with Heroku. BentoServices could quickly deploy to Heroku as API model server for production.

This guide demonstrates how to deploy a scikit-learn based iris classifier model with BentoML to Heroku. The same deployment steps are also applicable for models trained with other machine learning frameworks, see more BentoML examples here.


Heroku deployment with BentoML

Run the example project from the quick start guide to create the BentoML saved bundle for deployment:

git clone
pip install -r ./bentoml/guides/quick-start/requirements.txt
python ./bentoml/guides/quick-start/

Verify the saved bundle created:

$ bentoml get IrisClassifier:latest

# Sample output

  "name": "IrisClassifier",
  "version": "20200121141808_FE78B5",
  "uri": {
    "type": "LOCAL",
    "uri": "/Users/bozhaoyu/bentoml/repository/IrisClassifier/20200121141808_FE78B5"
  "bentoServiceMetadata": {
    "name": "IrisClassifier",
    "version": "20200121141808_FE78B5",
    "createdAt": "2020-01-21T22:18:25.079723Z",
    "env": {
      "condaEnv": "name: bentoml-IrisClassifier\nchannels:\n- defaults\ndependencies:\n- python=3.7.3\n- pip\n",
      "pipDependencies": "bentoml==0.5.8\nscikit-learn",
      "pythonVersion": "3.7.3"
    "artifacts": [
        "name": "model",
        "artifactType": "SklearnModelArtifact"
    "apis": [
        "name": "predict",
        "InputType": "DataframeInput",
        "docs": "BentoService API"

The BentoML saved bundle created can now be used to start a REST API Server hosting the BentoService and available for sending test request:

# Start BentoML API server:
bentoml serve IrisClassifier:latest
# Send test request:
curl -i \
  --header "Content-Type: application/json" \
  --request POST \
  --data '[[5.1, 3.5, 1.4, 0.2]]' \

Build and deploy to Heroku

Follow the CLI instruction and login to a Heroku account:

heroku login

Login to the Heroku Container Registry:

heroku container:login

Create a Heroku app:

APP_NAME=bentoml-her0ku-$(date +%s | base64 | tr '[:upper:]' '[:lower:]' | tr -dc _a-z-0-9)
heroku create $APP_NAME

Find the IrisClassifier SavedBundle directory:

cd $(bentoml get IrisClassifier:latest --print-location --quiet)

Build and push API server container with the SavedBundle, and push to the Heroku app bentoml-iris-classifier created above:

heroku container:push web --app $APP_NAME

Release the app:

heroku container:release web --app $APP_NAME

To view the deployment logs on heroku and verify the web server has been created:

heroku logs --tail -a $APP_NAME

Now, make prediction request with sample data:

curl -i \
  --header "Content-Type: application/json" \
  --request POST \
  --data '[[5.1, 3.5, 1.4, 0.2]]' \
  $(heroku apps:info --app $APP_NAME -j | jq -r ".app.web_url")/predict

Remove deployment on Heroku

heroku apps:destroy $APP_NAME