huggingface-hub 0.33.4__py3-none-any.whl → 0.34.0rc0__py3-none-any.whl

huggingface_hub/__init__.py

@@ -46,7 +46,7 @@ import sys
 from typing import TYPE_CHECKING
 
 
-__version__ = "0.33.4"
+__version__ = "0.34.0.rc0"
 
 # Alphabetical order of definitions is ensured in tests
 # WARNING: any comment added in this dictionary definition will be lost when
@@ -62,6 +62,12 @@ _SUBMOD_ATTRS = {
         "InferenceEndpointTimeoutError",
         "InferenceEndpointType",
     ],
+    "_jobs_api": [
+        "JobInfo",
+        "JobOwner",
+        "JobStage",
+        "JobStatus",
+    ],
     "_login": [
         "auth_list",
         "auth_switch",
@@ -165,6 +171,7 @@ _SUBMOD_ATTRS = {
         "add_space_variable",
         "auth_check",
         "cancel_access_request",
+        "cancel_job",
         "change_discussion_status",
         "comment_discussion",
         "create_branch",
@@ -194,6 +201,7 @@ _SUBMOD_ATTRS = {
         "duplicate_space",
         "edit_discussion_comment",
         "enable_webhook",
+        "fetch_job_logs",
         "file_exists",
         "get_collection",
         "get_dataset_tags",
@@ -210,11 +218,13 @@ _SUBMOD_ATTRS = {
         "get_user_overview",
         "get_webhook",
         "grant_access",
+        "inspect_job",
         "list_accepted_access_requests",
         "list_collections",
         "list_datasets",
         "list_inference_catalog",
         "list_inference_endpoints",
+        "list_jobs",
         "list_lfs_files",
         "list_liked_repos",
         "list_models",
@@ -251,6 +261,8 @@ _SUBMOD_ATTRS = {
         "resume_inference_endpoint",
         "revision_exists",
         "run_as_future",
+        "run_job",
+        "run_uv_job",
         "scale_to_zero_inference_endpoint",
         "set_space_sleep_time",
         "space_info",
@@ -360,6 +372,10 @@ _SUBMOD_ATTRS = {
         "ImageToTextInput",
         "ImageToTextOutput",
         "ImageToTextParameters",
+        "ImageToVideoInput",
+        "ImageToVideoOutput",
+        "ImageToVideoParameters",
+        "ImageToVideoTargetSize",
         "ObjectDetectionBoundingBox",
         "ObjectDetectionInput",
         "ObjectDetectionOutputElement",
@@ -648,6 +664,10 @@ __all__ = [
     "ImageToTextInput",
     "ImageToTextOutput",
     "ImageToTextParameters",
+    "ImageToVideoInput",
+    "ImageToVideoOutput",
+    "ImageToVideoParameters",
+    "ImageToVideoTargetSize",
     "InferenceApi",
     "InferenceClient",
     "InferenceEndpoint",
@@ -656,6 +676,10 @@ __all__ = [
     "InferenceEndpointTimeoutError",
     "InferenceEndpointType",
     "InferenceTimeoutError",
+    "JobInfo",
+    "JobOwner",
+    "JobStage",
+    "JobStatus",
     "KerasModelHubMixin",
     "MCPClient",
     "ModelCard",
@@ -792,6 +816,7 @@ __all__ = [
     "auth_switch",
     "cached_assets_path",
     "cancel_access_request",
+    "cancel_job",
     "change_discussion_status",
     "comment_discussion",
     "configure_http_backend",
@@ -825,6 +850,7 @@ __all__ = [
     "enable_webhook",
     "export_entries_as_dduf",
     "export_folder_as_dduf",
+    "fetch_job_logs",
     "file_exists",
     "from_pretrained_fastai",
     "from_pretrained_keras",
@@ -851,12 +877,14 @@ __all__ = [
     "grant_access",
     "hf_hub_download",
     "hf_hub_url",
+    "inspect_job",
     "interpreter_login",
     "list_accepted_access_requests",
     "list_collections",
     "list_datasets",
     "list_inference_catalog",
     "list_inference_endpoints",
+    "list_jobs",
     "list_lfs_files",
     "list_liked_repos",
     "list_models",
@@ -907,6 +935,8 @@ __all__ = [
     "resume_inference_endpoint",
     "revision_exists",
     "run_as_future",
+    "run_job",
+    "run_uv_job",
     "save_pretrained_keras",
     "save_torch_model",
     "save_torch_state_dict",
@@ -1044,6 +1074,12 @@ if TYPE_CHECKING:  # pragma: no cover
         InferenceEndpointTimeoutError,  # noqa: F401
         InferenceEndpointType,  # noqa: F401
     )
+    from ._jobs_api import (
+        JobInfo,  # noqa: F401
+        JobOwner,  # noqa: F401
+        JobStage,  # noqa: F401
+        JobStatus,  # noqa: F401
+    )
     from ._login import (
         auth_list,  # noqa: F401
         auth_switch,  # noqa: F401
@@ -1143,6 +1179,7 @@ if TYPE_CHECKING:  # pragma: no cover
         add_space_variable,  # noqa: F401
         auth_check,  # noqa: F401
         cancel_access_request,  # noqa: F401
+        cancel_job,  # noqa: F401
         change_discussion_status,  # noqa: F401
         comment_discussion,  # noqa: F401
         create_branch,  # noqa: F401
@@ -1172,6 +1209,7 @@ if TYPE_CHECKING:  # pragma: no cover
         duplicate_space,  # noqa: F401
         edit_discussion_comment,  # noqa: F401
         enable_webhook,  # noqa: F401
+        fetch_job_logs,  # noqa: F401
         file_exists,  # noqa: F401
         get_collection,  # noqa: F401
         get_dataset_tags,  # noqa: F401
@@ -1188,11 +1226,13 @@ if TYPE_CHECKING:  # pragma: no cover
         get_user_overview,  # noqa: F401
         get_webhook,  # noqa: F401
         grant_access,  # noqa: F401
+        inspect_job,  # noqa: F401
         list_accepted_access_requests,  # noqa: F401
         list_collections,  # noqa: F401
         list_datasets,  # noqa: F401
         list_inference_catalog,  # noqa: F401
         list_inference_endpoints,  # noqa: F401
+        list_jobs,  # noqa: F401
         list_lfs_files,  # noqa: F401
         list_liked_repos,  # noqa: F401
         list_models,  # noqa: F401
@@ -1229,6 +1269,8 @@ if TYPE_CHECKING:  # pragma: no cover
         resume_inference_endpoint,  # noqa: F401
         revision_exists,  # noqa: F401
         run_as_future,  # noqa: F401
+        run_job,  # noqa: F401
+        run_uv_job,  # noqa: F401
         scale_to_zero_inference_endpoint,  # noqa: F401
         set_space_sleep_time,  # noqa: F401
         space_info,  # noqa: F401
@@ -1336,6 +1378,10 @@ if TYPE_CHECKING:  # pragma: no cover
         ImageToTextInput,  # noqa: F401
         ImageToTextOutput,  # noqa: F401
         ImageToTextParameters,  # noqa: F401
+        ImageToVideoInput,  # noqa: F401
+        ImageToVideoOutput,  # noqa: F401
+        ImageToVideoParameters,  # noqa: F401
+        ImageToVideoTargetSize,  # noqa: F401
         ObjectDetectionBoundingBox,  # noqa: F401
         ObjectDetectionInput,  # noqa: F401
         ObjectDetectionOutputElement,  # noqa: F401

huggingface_hub/_commit_api.py

@@ -4,7 +4,6 @@ Type definitions and utilities for the `create_commit` API
 
 import base64
 import io
-import math
 import os
 import warnings
 from collections import defaultdict
@@ -23,6 +22,7 @@ from .lfs import UploadInfo, lfs_upload, post_lfs_batch_info
 from .utils import (
     FORBIDDEN_FOLDERS,
     XetTokenType,
+    are_progress_bars_disabled,
     chunk_iterable,
     fetch_xet_connection_info_from_repo_info,
     get_session,
@@ -33,7 +33,6 @@ from .utils import (
     validate_hf_hub_args,
 )
 from .utils import tqdm as hf_tqdm
-from .utils.tqdm import _get_progress_bar_context
 
 
 if TYPE_CHECKING:
@@ -529,9 +528,12 @@ def _upload_xet_files(
     """
     if len(additions) == 0:
         return
+
     # at this point, we know that hf_xet is installed
     from hf_xet import upload_bytes, upload_files
 
+    from .utils._xet_progress_reporting import XetProgressReporter
+
     try:
         xet_connection_info = fetch_xet_connection_info_from_repo_info(
             token_type=XetTokenType.WRITE,
@@ -567,32 +569,18 @@ def _upload_xet_files(
             raise XetRefreshTokenError("Failed to refresh xet token")
         return new_xet_connection.access_token, new_xet_connection.expiration_unix_epoch
 
-    num_chunks = math.ceil(len(additions) / UPLOAD_BATCH_MAX_NUM_FILES)
-    num_chunks_num_digits = int(math.log10(num_chunks)) + 1
-    for i, chunk in enumerate(chunk_iterable(additions, chunk_size=UPLOAD_BATCH_MAX_NUM_FILES)):
-        _chunk = [op for op in chunk]
-
-        bytes_ops = [op for op in _chunk if isinstance(op.path_or_fileobj, bytes)]
-        paths_ops = [op for op in _chunk if isinstance(op.path_or_fileobj, (str, Path))]
-        expected_size = sum(op.upload_info.size for op in bytes_ops + paths_ops)
+    if not are_progress_bars_disabled():
+        progress = XetProgressReporter()
+        progress_callback = progress.update_progress
+    else:
+        progress, progress_callback = None, None
 
-        if num_chunks > 1:
-            description = f"Uploading Batch [{str(i + 1).zfill(num_chunks_num_digits)}/{num_chunks}]..."
-        else:
-            description = "Uploading..."
-        progress_cm = _get_progress_bar_context(
-            desc=description,
-            total=expected_size,
-            initial=0,
-            unit="B",
-            unit_scale=True,
-            name="huggingface_hub.xet_put",
-            log_level=logger.getEffectiveLevel(),
-        )
-        with progress_cm as progress:
+    try:
+        for i, chunk in enumerate(chunk_iterable(additions, chunk_size=UPLOAD_BATCH_MAX_NUM_FILES)):
+            _chunk = [op for op in chunk]
 
-            def update_progress(increment: int):
-                progress.update(increment)
+            bytes_ops = [op for op in _chunk if isinstance(op.path_or_fileobj, bytes)]
+            paths_ops = [op for op in _chunk if isinstance(op.path_or_fileobj, (str, Path))]
 
             if len(paths_ops) > 0:
                 upload_files(
@@ -600,7 +588,7 @@ def _upload_xet_files(
                     xet_endpoint,
                     access_token_info,
                     token_refresher,
-                    update_progress,
+                    progress_callback,
                     repo_type,
                 )
             if len(bytes_ops) > 0:
@@ -609,9 +597,14 @@ def _upload_xet_files(
                     xet_endpoint,
                     access_token_info,
                     token_refresher,
-                    update_progress,
+                    progress_callback,
                     repo_type,
                 )
+
+    finally:
+        if progress is not None:
+            progress.close(False)
+
     return
 
 

huggingface_hub/_jobs_api.py

@@ -0,0 +1,145 @@
+# coding=utf-8
+# Copyright 2025-present, the HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+from dataclasses import dataclass
+from datetime import datetime
+from enum import Enum
+from typing import Any, Dict, List, Optional
+
+from huggingface_hub import constants
+from huggingface_hub._space_api import SpaceHardware
+from huggingface_hub.utils._datetime import parse_datetime
+
+
+class JobStage(str, Enum):
+    """
+    Enumeration of possible stage of a Job on the Hub.
+
+    Value can be compared to a string:
+    ```py
+    assert JobStage.COMPLETED == "COMPLETED"
+    ```
+
+    Taken from https://github.com/huggingface/moon-landing/blob/main/server/job_types/JobInfo.ts#L61 (private url).
+    """
+
+    # Copied from moon-landing > server > lib > Job.ts
+    COMPLETED = "COMPLETED"
+    CANCELED = "CANCELED"
+    ERROR = "ERROR"
+    DELETED = "DELETED"
+    RUNNING = "RUNNING"
+
+
+@dataclass
+class JobStatus:
+    stage: JobStage
+    message: Optional[str]
+
+    def __init__(self, **kwargs) -> None:
+        self.stage = kwargs["stage"]
+        self.message = kwargs.get("message")
+
+
+@dataclass
+class JobOwner:
+    id: str
+    name: str
+
+
+@dataclass
+class JobInfo:
+    """
+    Contains information about a Job.
+
+    Args:
+        id (`str`):
+            Job ID.
+        created_at (`datetime` or `None`):
+            When the Job was created.
+        docker_image (`str` or `None`):
+            The Docker image from Docker Hub used for the Job.
+            Can be None if space_id is present instead.
+        space_id (`str` or `None`):
+            The Docker image from Hugging Face Spaces used for the Job.
+            Can be None if docker_image is present instead.
+        command (`List[str]` or `None`):
+            Command of the Job, e.g. `["python", "-c", "print('hello world')"]`
+        arguments (`List[str]` or `None`):
+            Arguments passed to the command
+        environment (`Dict[str]` or `None`):
+            Environment variables of the Job as a dictionary.
+        secrets (`Dict[str]` or `None`):
+            Secret environment variables of the Job (encrypted).
+        flavor (`str` or `None`):
+            Flavor for the hardware, as in Hugging Face Spaces. See [`SpaceHardware`] for possible values.
+            E.g. `"cpu-basic"`.
+        status: (`JobStatus` or `None`):
+            Status of the Job, e.g. `JobStatus(stage="RUNNING", message=None)`
+            See [`JobStage`] for possible stage values.
+        status: (`JobOwner` or `None`):
+            Owner of the Job, e.g. `JobOwner(id="5e9ecfc04957053f60648a3e", name="lhoestq")`
+
+    Example:
+
+    ```python
+    >>> from huggingface_hub import run_job
+    >>> job = run_job(
+    ...     image="python:3.12",
+    ...     command=["python", "-c", "print('Hello from the cloud!')"]
+    ... )
+    >>> job
+    JobInfo(id='687fb701029421ae5549d998', created_at=datetime.datetime(2025, 7, 22, 16, 6, 25, 79000, tzinfo=datetime.timezone.utc), docker_image='python:3.12', space_id=None, command=['python', '-c', "print('Hello from the cloud!')"], arguments=[], environment={}, secrets={}, flavor='cpu-basic', status=JobStatus(stage='RUNNING', message=None), owner=JobOwner(id='5e9ecfc04957053f60648a3e', name='lhoestq'), endpoint='https://huggingface.co', url='https://huggingface.co/jobs/lhoestq/687fb701029421ae5549d998')
+    >>> job.id
+    '687fb701029421ae5549d998'
+    >>> job.url
+    'https://huggingface.co/jobs/lhoestq/687fb701029421ae5549d998'
+    >>> job.status.stage
+    'RUNNING'
+    ```
+    """
+
+    id: str
+    created_at: Optional[datetime]
+    docker_image: Optional[str]
+    space_id: Optional[str]
+    command: Optional[List[str]]
+    arguments: Optional[List[str]]
+    environment: Optional[Dict[str, Any]]
+    secrets: Optional[Dict[str, Any]]
+    flavor: Optional[SpaceHardware]
+    status: Optional[JobStatus]
+    owner: Optional[JobOwner]
+
+    # Inferred fields
+    endpoint: str
+    url: str
+
+    def __init__(self, **kwargs) -> None:
+        self.id = kwargs["id"]
+        created_at = kwargs.get("createdAt") or kwargs.get("created_at")
+        self.created_at = parse_datetime(created_at) if created_at else None
+        self.docker_image = kwargs.get("dockerImage") or kwargs.get("docker_image")
+        self.space_id = kwargs.get("spaceId") or kwargs.get("space_id")
+        self.owner = JobOwner(**(kwargs["owner"] if isinstance(kwargs.get("owner"), dict) else {}))
+        self.command = kwargs.get("command")
+        self.arguments = kwargs.get("arguments")
+        self.environment = kwargs.get("environment")
+        self.secrets = kwargs.get("secrets")
+        self.flavor = kwargs.get("flavor")
+        self.status = JobStatus(**(kwargs["status"] if isinstance(kwargs.get("status"), dict) else {}))
+
+        # Inferred fields
+        self.endpoint = kwargs.get("endpoint", constants.ENDPOINT)
+        self.url = f"{self.endpoint}/jobs/{self.owner.name}/{self.id}"

huggingface_hub/_local_folder.py

@@ -86,7 +86,13 @@ class LocalDownloadFilePaths:
 
     def incomplete_path(self, etag: str) -> Path:
         """Return the path where a file will be temporarily downloaded before being moved to `file_path`."""
-        return self.metadata_path.parent / f"{_short_hash(self.metadata_path.name)}.{etag}.incomplete"
+        path = self.metadata_path.parent / f"{_short_hash(self.metadata_path.name)}.{etag}.incomplete"
+        resolved_path = str(path.resolve())
+        # Some Windows versions do not allow for paths longer than 255 characters.
+        # In this case, we must specify it as an extended path by using the "\\?\" prefix.
+        if len(resolved_path) > 255 and not resolved_path.startswith("\\\\?\\"):
+            path = Path("\\\\?\\" + resolved_path)
+        return path
 
 
 @dataclass(frozen=True)

huggingface_hub/_login.py

@@ -75,7 +75,7 @@ def login(
     components. If `token` is not provided, it will be prompted to the user either with
     a widget (in a notebook) or via the terminal.
 
-    To log in from outside of a script, one can also use `huggingface-cli login` which is
+    To log in from outside of a script, one can also use `hf auth login` which is
     a cli command that wraps [`login`].
 
     <Tip>
@@ -120,7 +120,7 @@ def login(
             logger.info(
                 "The token has not been saved to the git credentials helper. Pass "
                 "`add_to_git_credential=True` in this function directly or "
-                "`--add-to-git-credential` if using via `huggingface-cli` if "
+                "`--add-to-git-credential` if using via `hf`CLI if "
                 "you want to set the git credential as well."
             )
         _login(token, add_to_git_credential=add_to_git_credential)
@@ -233,7 +233,7 @@ def auth_list() -> None:
         )
     elif current_token_name is None:
         logger.warning(
-            "\nNote: No active token is set and no environment variable `HF_TOKEN` is found. Use `huggingface-cli login` to log in."
+            "\nNote: No active token is set and no environment variable `HF_TOKEN` is found. Use `hf auth login` to log in."
         )
 
 
@@ -273,8 +273,8 @@ def interpreter_login(*, new_session: bool = True, write_permission: bool = Fals
     print(_HF_LOGO_ASCII)
     if get_token() is not None:
         logger.info(
-            "    A token is already saved on your machine. Run `huggingface-cli"
-            " whoami` to get more information or `huggingface-cli logout` if you want"
+            "    A token is already saved on your machine. Run `hf auth whoami`"
+            " to get more information or `hf auth logout` if you want"
             " to log out."
         )
         logger.info("    Setting a new token will erase the existing one.")

huggingface_hub/_oauth.py

@@ -415,7 +415,7 @@ def _get_mocked_oauth_info() -> Dict:
     if token is None:
         raise ValueError(
             "Your machine must be logged in to HF to debug an OAuth app locally. Please"
-            " run `huggingface-cli login` or set `HF_TOKEN` as environment variable "
+            " run `hf auth login` or set `HF_TOKEN` as environment variable "
             "with one of your access token. You can generate a new token in your "
             "settings page (https://huggingface.co/settings/tokens)."
         )

huggingface_hub/_snapshot_download.py

@@ -254,14 +254,19 @@ def snapshot_download(
     # At this stage, internet connection is up and running
     # => let's download the files!
     assert repo_info.sha is not None, "Repo info returned from server must have a revision sha."
-    assert repo_info.siblings is not None, "Repo info returned from server must have a siblings list."
 
     # Corner case: on very large repos, the siblings list in `repo_info` might not contain all files.
     # In that case, we need to use the `list_repo_tree` method to prevent caching issues.
-    repo_files: Iterable[str] = [f.rfilename for f in repo_info.siblings]
-    has_many_files = len(repo_info.siblings) > VERY_LARGE_REPO_THRESHOLD
-    if has_many_files:
-        logger.info("The repo has more than 50,000 files. Using `list_repo_tree` to ensure all files are listed.")
+    repo_files: Iterable[str] = [f.rfilename for f in repo_info.siblings] if repo_info.siblings is not None else []
+    unreliable_nb_files = (
+        repo_info.siblings is None
+        or len(repo_info.siblings) == 0
+        or len(repo_info.siblings) > VERY_LARGE_REPO_THRESHOLD
+    )
+    if unreliable_nb_files:
+        logger.info(
+            "Number of files in the repo is unreliable. Using `list_repo_tree` to ensure all files are listed."
+        )
         repo_files = (
             f.rfilename
             for f in api.list_repo_tree(repo_id=repo_id, recursive=True, revision=revision, repo_type=repo_type)
@@ -274,7 +279,7 @@ def snapshot_download(
         ignore_patterns=ignore_patterns,
     )
 
-    if not has_many_files:
+    if not unreliable_nb_files:
         filtered_repo_files = list(filtered_repo_files)
         tqdm_desc = f"Fetching {len(filtered_repo_files)} files"
     else: