inference

models.pull

Download an inference model locally.

strata inference models pull

wire inference_models_pull

Resolves a catalog name or model spec and downloads the model artifact into the local model directory, returning the resolved local path. A model that is already present is not downloaded again: the command returns its path in every build, with or without network access. Honors `STRATA_MODELS_DIR` for the destination and `STRATA_HF_ENDPOINT` and `STRATA_HF_TOKEN` (or `HF_TOKEN`) for gated HuggingFace repositories. The spec is resolved before anything else is checked: a malformed spec returns `inference.invalid_request`, a cloud provider spec returns `inference.unsupported_operation` (there is nothing to download), and a name that is not in the catalog returns `inference.missing_model`. A missing model is downloaded only when the runtime has network access and the build can download; otherwise the command returns `inference.download_disabled`.

To see the complete help for this command, run:

strata inference models pull --help

Parameters

NameTypeRequiredDescription
modelstringyesModel spec or catalog name.

Returns

PullModelOutput

  • dataPullModelOutput
  • typestring

Errors

Recover by code. Retry policy and commit outcome come from the shipped error registry.

CodeRetryCommit outcome
failed_precondition.engine.runtime_closednevernot_started
not_found.engine.branchnevernot_applicable
inference.invalid_requestnevernot_applicable
inference.unsupported_operationnevernot_applicable
inference.download_disabledafter_state_changenot_applicable
inference.missing_modelafter_state_changenot_applicable
inference.unknown_modelnevernot_applicable
inference.download_failedsame_requestnot_applicable
inference.download_verification_failedafter_state_changenot_applicable
inference.io_failureafter_state_changenot_applicable
inference.registry_corruptnevernot_applicable
← All 12 Inference commands

agents: this page as markdown → /docs/reference/inference/models/pull.md