CLI Reference

bentoml

BentoML CLI

bentoml [OPTIONS] COMMAND [ARGS]...

Options

--version

Show the version and exit.

build

Build a new Bento from current directory

bentoml build [OPTIONS] [BUILD_CTX]

Options

-f, --bentofile <bentofile>
--version <version>
-q, --quiet

Suppress all warnings and info logs

--verbose, --debug

Generate debug information

--do-not-track

Do not send usage info

--config <config>

BentoML configuration YAML file to apply

Arguments

BUILD_CTX

Optional argument

Environment variables

BENTOML_DO_NOT_TRACK

Provide a default for --do-not-track

BENTOML_CONFIG

Provide a default for --config

containerize

Containerizes given Bento into a ready-to-use Docker image.

bentoml containerize [OPTIONS] BENTO_TAG

Options

-t, --docker-image-tag <docker_image_tag>

Docker image tag, default to same as the Bento tag

--build-arg <build_arg>

docker image build args

--label <label>

docker image label

--no-cache
--platform <platform>
-q, --quiet

Suppress all warnings and info logs

--verbose, --debug

Generate debug information

--do-not-track

Do not send usage info

--config <config>

BentoML configuration YAML file to apply

Arguments

BENTO_TAG

Required argument

Environment variables

BENTOML_DO_NOT_TRACK

Provide a default for --do-not-track

BENTOML_CONFIG

Provide a default for --config

delete

Delete Bento in local bento store.

Specify target Bentos to remove:

  • Delete single bento bundle by “name:version”, e.g: bentoml delete IrisClassifier:v1

  • Bulk delete all bento bundles with a specific name, e.g.: bentoml delete IrisClassifier

  • Bulk delete multiple bento bundles by name and version, separated by “,”, e.g.: benotml delete Irisclassifier:v1,MyPredictService:v2

bentoml delete [OPTIONS] DELETE_TARGETS

Options

-y, --yes, --assume-yes

Skip confirmation when deleting a specific bento bundle

-q, --quiet

Suppress all warnings and info logs

--verbose, --debug

Generate debug information

--do-not-track

Do not send usage info

--config <config>

BentoML configuration YAML file to apply

Arguments

DELETE_TARGETS

Required argument

Environment variables

BENTOML_DO_NOT_TRACK

Provide a default for --do-not-track

BENTOML_CONFIG

Provide a default for --config

export

Export Bento to a tar file

bentoml export [OPTIONS] BENTO_TAG [OUT_FILE]

Options

-q, --quiet

Suppress all warnings and info logs

--verbose, --debug

Generate debug information

--do-not-track

Do not send usage info

--config <config>

BentoML configuration YAML file to apply

Arguments

BENTO_TAG

Required argument

OUT_FILE

Optional argument

Environment variables

BENTOML_DO_NOT_TRACK

Provide a default for --do-not-track

BENTOML_CONFIG

Provide a default for --config

get

Get Bento information

bentoml get [OPTIONS] BENTO_TAG

Options

-o, --output <output>
Options

tree | json | yaml | path

-q, --quiet

Suppress all warnings and info logs

--verbose, --debug

Generate debug information

--do-not-track

Do not send usage info

--config <config>

BentoML configuration YAML file to apply

Arguments

BENTO_TAG

Required argument

Environment variables

BENTOML_DO_NOT_TRACK

Provide a default for --do-not-track

BENTOML_CONFIG

Provide a default for --config

import

Import a previously exported Bento tar file

bentoml import [OPTIONS] [BENTO_PATH]

Options

-q, --quiet

Suppress all warnings and info logs

--verbose, --debug

Generate debug information

--do-not-track

Do not send usage info

--config <config>

BentoML configuration YAML file to apply

Arguments

BENTO_PATH

Optional argument

Environment variables

BENTOML_DO_NOT_TRACK

Provide a default for --do-not-track

BENTOML_CONFIG

Provide a default for --config

list

List Bentos in local bento store

bentoml list [OPTIONS] [BENTO_NAME]

Options

-o, --output <output>
Options

json | yaml | table

--no-trunc <no_trunc>

Don’t truncate the output

-q, --quiet

Suppress all warnings and info logs

--verbose, --debug

Generate debug information

--do-not-track

Do not send usage info

--config <config>

BentoML configuration YAML file to apply

Arguments

BENTO_NAME

Optional argument

Environment variables

BENTOML_DO_NOT_TRACK

Provide a default for --do-not-track

BENTOML_CONFIG

Provide a default for --config

models

Model Management

bentoml models [OPTIONS] COMMAND [ARGS]...

delete

Delete Model in local model store.

Specify target Models to remove:

  • Delete single model by “name:version”, e.g: bentoml model delete IrisClassifier:v1

  • Bulk delete all models with a specific name, e.g.: bentoml model delete IrisClassifier

  • Bulk delete multiple models by name and version, separated by “,”, e.g.: benotml model delete Irisclassifier:v1,MyPredictService:v2

bentoml models delete [OPTIONS] DELETE_TARGETS

Options

-y, --yes, --assume-yes

Skip confirmation when deleting a specific model

Arguments

DELETE_TARGETS

Required argument

export

Export Model to a tar file

bentoml models export [OPTIONS] MODEL_TAG [OUT_FILE]

Arguments

MODEL_TAG

Required argument

OUT_FILE

Optional argument

get

Get Model information

bentoml models get [OPTIONS] MODEL_TAG

Options

-o, --output <output>
Options

tree | json | yaml | path

Arguments

MODEL_TAG

Required argument

import

Import a previously exported Model tar file

bentoml models import [OPTIONS] [MODEL_PATH]

Arguments

MODEL_PATH

Optional argument

list

List Models in local model store

bentoml models list [OPTIONS] [MODEL_NAME]

Options

-o, --output <output>
Options

json | yaml | table

--no-trunc <no_trunc>

Don’t truncate the output

Arguments

MODEL_NAME

Optional argument

pull

Pull Model from a yatai server

bentoml models pull [OPTIONS] MODEL_TAG

Options

-f, --force

Force pull from yatai to local and overwrite even if it already exists in local

Arguments

MODEL_TAG

Required argument

push

Push Model to a yatai server

bentoml models push [OPTIONS] MODEL_TAG

Options

-f, --force

Forced push to yatai even if it exists in yatai

Arguments

MODEL_TAG

Required argument

pull

Pull Bento from a yatai server

bentoml pull [OPTIONS] BENTO_TAG

Options

-f, --force

Force pull from yatai to local and overwrite even if it already exists in local

-q, --quiet

Suppress all warnings and info logs

--verbose, --debug

Generate debug information

--do-not-track

Do not send usage info

--config <config>

BentoML configuration YAML file to apply

Arguments

BENTO_TAG

Required argument

Environment variables

BENTOML_DO_NOT_TRACK

Provide a default for --do-not-track

BENTOML_CONFIG

Provide a default for --config

push

Push Bento to a yatai server

bentoml push [OPTIONS] BENTO_TAG

Options

-f, --force

Forced push to yatai even if it exists in yatai

-q, --quiet

Suppress all warnings and info logs

--verbose, --debug

Generate debug information

--do-not-track

Do not send usage info

--config <config>

BentoML configuration YAML file to apply

Arguments

BENTO_TAG

Required argument

Environment variables

BENTOML_DO_NOT_TRACK

Provide a default for --do-not-track

BENTOML_CONFIG

Provide a default for --config

serve

Start BentoServer from BENTO

BENTO is the serving target: it can be the import path of a bentoml.Service instance; a tag to a Bento in local Bento store; or a file path to a Bento directory, e.g.:

Serve from a bentoml.Service instance source code(for development use only):
bentoml serve fraud_detector.py:svc
Serve from a Bento built in local store:
bentoml serve fraud_detector:4tht2icroji6zput3suqi5nl2
bentoml serve fraud_detector:latest
Serve from a Bento directory:
bentoml serve ./fraud_detector_bento
bentoml serve [OPTIONS] BENTO

Options

--production

Run the BentoServer in production mode

Default

False

--port <port>

The port to listen on for the REST api server

Default

5000

--host <host>

The host to bind for the REST api server [defaults: 127.0.0.1(dev), 0.0.0.0(production)]

--backlog <backlog>

The maximum number of pending connections.

Default

2048

--reload

Reload Service when code changes detected, this is only available in development mode

Default

False

--reload-delay <reload_delay>

Delay in seconds between each check if the Service needs to be reloaded

Default

0.25

--working-dir <working_dir>

When loading from source code, specify the directory to find the Service instance

Default

.

--run-with-ngrok, --ngrok

Use ngrok to relay traffic on a public endpoint to the local BentoServer, only available in dev mode

Default

False

-q, --quiet

Suppress all warnings and info logs

--verbose, --debug

Generate debug information

--do-not-track

Do not send usage info

--config <config>

BentoML configuration YAML file to apply

Arguments

BENTO

Required argument

Environment variables

BENTOML_PORT

Provide a default for --port

BENTOML_HOST

Provide a default for --host

BENTOML_DO_NOT_TRACK

Provide a default for --do-not-track

BENTOML_CONFIG

Provide a default for --config

yatai

yatai sub commands

bentoml yatai [OPTIONS] COMMAND [ARGS]...

login

Login to yatai server

bentoml yatai login [OPTIONS]

Options

--endpoint <endpoint>

Yatai endpoint, like https://yatai.ai

--api-token <api_token>

Yatai user api token