baseten model

Manage Baseten models

Create, list, and push Baseten models.

Authentication is through ‘baseten auth login’ or the BASETEN_API_KEY environment variable.

push

baseten model push [OPTIONS] [--dir DIR]

Build a model archive, upload it to Baseten, and create either a new model or a new deployment of an existing model.

The current directory is used by default; pass --dir to push a model directory at another path.

The model is identified by the model_name field in config.yaml. Use --override-name to override that for this push only.

Options

--deploy-timeout · TEXT

Deployment timeout as a Go duration (e.g. 30m, 1h); allowed range 10m to 24h.

--deployment-name · TEXT

Human-readable name for the new deployment.

--develop · BOOL

Push as a development deployment: the model’s single mutable dev slot, created if absent and overwritten in place otherwise. Incompatible with —environment and —deployment-name.

--dir · TEXT · default: "."

Model directory to push. Defaults to the current directory.

--disable-archive-download · BOOL

Disable archive download for the new model. Only valid for new models.

--dry-run · BOOL

Validate the push and request upload credentials without uploading or creating anything.

--environment · TEXT

Stable environment to push to.

-q, --jq · TEXT

Filter JSON output with a jq expression; implies —output json (or jsonl for streamed commands)

--labels · TEXT

User-provided labels for the deployment as a JSON object, e.g. ’{“team”:“ml”,“priority”:1}’.

--no-build-cache · BOOL

Force a full rebuild without using cached layers.

-o, --output · TEXT · default: "text"

Output format

One of: text, json, jsonl, none

--override-env-instance-type · BOOL

Use this deployment’s instance type instead of preserving the target environment’s. Only meaningful when an environment is targeted.

--override-name · TEXT

Override the model_name from config.yaml for this push only. The on-disk config.yaml is not modified.

--profile · TEXT

Use a specific stored profile for this command, overriding BASETEN_PROFILE and the current profile

--tail · BOOL

Stream build and runtime logs to stderr after pushing. Logs are always text-formatted; use ‘baseten model deployment logs —tail’ for structured log streaming.

--team · TEXT

Team the model belongs to. Only valid for new models.

--wait · BOOL

Block until the deployment is active. Exits non-zero on a terminal-failure status.

--watch · BOOL

After pushing, watch the model directory and live-patch the development deployment on change. Implies —develop.

--watch-hot-reload · BOOL

With —watch, hot-reload the running container when every change is to model code; mixed changes fall back to a cold patch.

--watch-no-keepalive · BOOL

With —watch, let the development deployment scale to zero while watching. By default it is kept warm by periodic pings.

-v, --verbose · BOOL

Enable verbose logging

Examples

Push the current directory as a new deployment

baseten model push

Push and stream build/runtime logs until the deployment is active

baseten model push --tail --wait

Filter output with --jq

Print the new deployment’s predict URL

baseten model push --jq '.predict_url'

Output

Text mode (--output text): Narrative summary on stdout: success banner, deployment status, log/predict URLs and example next-step commands. Under --output json the narrative is redirected to stderr so stdout stays a clean JSON document.

JSON mode (--output json): payload type cmd.ModelPushResult.

Under --dry-run no upload or deployment happens; the push is validated, upload credentials are requested, and stdout is the empty JSON object \{\}. Otherwise stdout is the full model+deployment result.

watch

baseten model watch [--dir DIR]

Watch a model directory and live-patch its development deployment. It applies each change as you save and runs until you interrupt it.

Options

--dir · TEXT · default: "."

Model directory to watch. Defaults to the current directory.

--hot-reload · BOOL

Hot-reload the running container when every change is to model code; mixed changes fall back to a cold patch.

-q, --jq · TEXT

Filter JSON output with a jq expression; implies —output json (or jsonl for streamed commands)

--no-keepalive · BOOL

Let the development deployment scale to zero while watching. By default it is kept warm by periodic pings.

-o, --output · TEXT · default: "text"

Output format

One of: text, json, jsonl, none

--profile · TEXT

Use a specific stored profile for this command, overriding BASETEN_PROFILE and the current profile

--team · TEXT

Team the model belongs to. Use to disambiguate when the same model_name exists in multiple teams.

-v, --verbose · BOOL

Enable verbose logging

Examples

Watch the current directory against its model’s development deployment

baseten model watch

Watch another directory and hot-reload on model-code changes

baseten model watch --dir ./my-model --hot-reload

Output

Text mode (--output text): Streams patch and sync status to stderr as changes are applied. Produces no stdout output.

list

baseten model list [OPTIONS]

List Baseten models.

Options

-q, --jq · TEXT

Filter JSON output with a jq expression; implies —output json (or jsonl for streamed commands)

-o, --output · TEXT · default: "text"

Output format

One of: text, json, jsonl, none

--profile · TEXT

Use a specific stored profile for this command, overriding BASETEN_PROFILE and the current profile

--team · TEXT

Team name or ID to scope the listing to. Defaults to all teams the caller can see.

-v, --verbose · BOOL

Enable verbose logging

Examples

List all models accessible to the caller

baseten model list

List only models in a specific team

baseten model list --team my-team

Filter output with --jq

Print just the model IDs

baseten model list --jq '.models[].id'

Output

Text mode (--output text): Table with columns: ID, NAME, TEAM, DEPLOYMENTS, CREATED. When no models exist, prints “No models found.” to stderr.

JSON mode (--output json): payload type managementapi.Models.

describe

baseten model describe [OPTIONS]

Describe a Baseten model.

Options

-q, --jq · TEXT

Filter JSON output with a jq expression; implies —output json (or jsonl for streamed commands)

--model-id · TEXT

ID of the model.

Mutually exclusive with other flags in group model-ref.

--model-name · TEXT

Name of the model. Use —team to disambiguate when the same name exists in multiple teams.

Mutually exclusive with other flags in group model-ref.

-o, --output · TEXT · default: "text"

Output format

One of: text, json, jsonl, none

--profile · TEXT

Use a specific stored profile for this command, overriding BASETEN_PROFILE and the current profile

--team · TEXT

Team name or ID. Only valid with —model-name.

-v, --verbose · BOOL

Enable verbose logging

Examples

Describe a model by ID

baseten model describe --model-id <model-id>

Describe a model by name

baseten model describe --model-name <name>

Filter output with --jq

Print the production deployment ID

baseten model describe --model-id <model-id> --jq '.production_deployment_id'

Output

Text mode (--output text): Field-per-line summary: ID, Name, Team, Deployments, Instance, Production, Development, Created. Optional fields are omitted when unset.

JSON mode (--output json): payload type managementapi.Model.

predict

baseten model predict [OPTIONS]

POST a JSON request to a model and write the response to stdout.

Targets the production environment by default. Use --environment, --deployment-id, or --regional to target something else.

Streaming responses (Transfer-Encoding: chunked) are passed through as they arrive. For machine-readable streaming JSON from OpenAI-compatible models, use --output jsonl.

Options

--data · TEXT

Inline JSON request body.

Mutually exclusive with other flags in group predict-input.

--deployment-id · TEXT

Specific deployment to target. Mutually exclusive with —environment and —regional.

--environment · TEXT

Environment to target (e.g. production, development). Defaults to production. Mutually exclusive with —deployment-id and —regional.

--file · TEXT

Path to a JSON file containing the request body. Use ’-’ for stdin.

Mutually exclusive with other flags in group predict-input.

-q, --jq · TEXT

Filter JSON output with a jq expression; implies —output json (or jsonl for streamed commands)

--model-id · TEXT

ID of the model.

Mutually exclusive with other flags in group model-ref.

--model-name · TEXT

Name of the model. Use —team to disambiguate when the same name exists in multiple teams.

Mutually exclusive with other flags in group model-ref.

-o, --output · TEXT · default: "text"

Output format

One of: text, json, jsonl, none

--profile · TEXT

Use a specific stored profile for this command, overriding BASETEN_PROFILE and the current profile

--regional · TEXT

Regional environment name; routes through the regional hostname. Mutually exclusive with —environment and —deployment-id.

--team · TEXT

Team name or ID. Only valid with —model-name.

--websocket · BOOL

Use the WebSocket predict endpoint. Sends the body as one frame, reads one frame back, then closes. Not for multi-message or back-and-forth sessions.

-v, --verbose · BOOL

Enable verbose logging

Examples

Send an inline JSON body

baseten model predict --model-id <model-id> --data '{"prompt":"hello"}'

Send a request body from a file

baseten model predict --model-id <model-id> --file request.json

Filter output with --jq

Extract a field when the model returns JSON

baseten model predict --model-id <model-id> --data '{"x":1}' --jq '.result'

Output

Text mode (--output text): The model’s response body, passed through verbatim. May be JSON, plain text, or binary, and may stream when the model uses chunked transfer encoding or SSE.

JSON mode (--output json): payload type cmd.JSONUndefined.

Under --output json, binary frames are base64-encoded under a ‘body’ key. Under --output jsonl, each SSE or binary chunk is emitted as its own record, one per line.

delete

baseten model delete [OPTIONS]

Delete a Baseten model and all of its deployments.

Prompts for the model name to confirm the deletion. Pass --yes to skip the prompt. When stdin is not a terminal, --yes is required.

Options

-q, --jq · TEXT

Filter JSON output with a jq expression; implies —output json (or jsonl for streamed commands)

--model-id · TEXT

ID of the model.

Mutually exclusive with other flags in group model-ref.

--model-name · TEXT

Name of the model. Use —team to disambiguate when the same name exists in multiple teams.

Mutually exclusive with other flags in group model-ref.

-o, --output · TEXT · default: "text"

Output format

One of: text, json, jsonl, none

--profile · TEXT

Use a specific stored profile for this command, overriding BASETEN_PROFILE and the current profile

--team · TEXT

Team name or ID. Only valid with —model-name.

--yes · BOOL

Skip the interactive confirmation prompt. Required when stdin is not a terminal.

-v, --verbose · BOOL

Enable verbose logging

Examples

Delete by ID without confirmation

baseten model delete --model-id <model-id> --yes

Delete by name with interactive confirmation

baseten model delete --model-name <name>

Filter output with --jq

Print the deleted model’s ID

baseten model delete --model-id <model-id> --yes --jq '.id'

Output

Text mode (--output text): On success, prints “Deleted model name (id)” to stderr; no stdout output.

JSON mode (--output json): payload type managementapi.ModelTombstone.