extract-python 0.10.3__tar.gz → 0.11.0__tar.gz

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (21) hide show
  1. {extract_python-0.10.3 → extract_python-0.11.0}/PKG-INFO +3 -1
  2. {extract_python-0.10.3 → extract_python-0.11.0}/extract_python/__init__.py +3 -3
  3. extract_python-0.11.0/extract_python/__main__.py +9 -0
  4. extract_python-0.11.0/extract_python/cli/__init__.py +50 -0
  5. extract_python-0.11.0/extract_python/cli/doc.py +78 -0
  6. extract_python-0.11.0/extract_python/cli/utils.py +25 -0
  7. {extract_python-0.10.3 → extract_python-0.11.0}/extract_python/docling_.py +1 -2
  8. {extract_python-0.10.3 → extract_python-0.11.0}/pyproject.toml +10 -0
  9. {extract_python-0.10.3 → extract_python-0.11.0}/uv.lock +280 -49
  10. {extract_python-0.10.3 → extract_python-0.11.0}/.gitignore +0 -0
  11. {extract_python-0.10.3 → extract_python-0.11.0}/.python-version +0 -0
  12. {extract_python-0.10.3 → extract_python-0.11.0}/README.md +0 -0
  13. {extract_python-0.10.3 → extract_python-0.11.0}/benches/__init__.py +0 -0
  14. {extract_python-0.10.3 → extract_python-0.11.0}/benches/compare.ipynb +0 -0
  15. {extract_python-0.10.3 → extract_python-0.11.0}/benches/compare.py +0 -0
  16. {extract_python-0.10.3 → extract_python-0.11.0}/benches/constants.py +0 -0
  17. {extract_python-0.10.3 → extract_python-0.11.0}/data/.gitignore +0 -0
  18. {extract_python-0.10.3 → extract_python-0.11.0}/extract_python/constants.py +0 -0
  19. {extract_python-0.10.3 → extract_python-0.11.0}/extract_python/marker_.py +0 -0
  20. {extract_python-0.10.3 → extract_python-0.11.0}/extract_python/miner_u.py +0 -0
  21. {extract_python-0.10.3 → extract_python-0.11.0}/extract_python/utils.py +0 -0
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.5
2
2
  Name: extract-python
3
- Version: 0.10.3
3
+ Version: 0.11.0
4
4
  Summary: Structured content extraction
5
5
  Project-URL: Homepage, https://github.com/ICIJ/extract-python
6
6
  Project-URL: Repository, https://github.com/ICIJ/extract-python
@@ -15,6 +15,8 @@ Requires-Dist: html2image~=2.0.7; extra == 'benches'
15
15
  Requires-Dist: markdown2>=2.5.4; extra == 'benches'
16
16
  Requires-Dist: notebook>=7.4.5; extra == 'benches'
17
17
  Requires-Dist: pypdfium2>=4.30.0; extra == 'benches'
18
+ Provides-Extra: cli
19
+ Requires-Dist: typer~=0.21.2; extra == 'cli'
18
20
  Provides-Extra: docling
19
21
  Requires-Dist: docling-slim[feat-ocr-easyocr,feat-ocr-mac,feat-ocr-rapidocr,feat-ocr-tesserocr,standard]~=2.127; extra == 'docling'
20
22
  Provides-Extra: docling-gpu
@@ -5,20 +5,20 @@ logger = logging.getLogger(__name__)
5
5
  try:
6
6
  from .docling_ import DOCLING_DEFAULT_ARTIFACTS_PATH, DoclingPipeline
7
7
  except ModuleNotFoundError:
8
- logger.exception("docling is not available, might be optional...")
8
+ logger.warning("docling is not available, might be optional...")
9
9
  DOCKING_DEFAULT_ARTIFACTS_PATH, DoclingPipeline = None, None
10
10
 
11
11
  try:
12
12
  from .marker_ import MarkerPipeline
13
13
  except ModuleNotFoundError:
14
- logger.exception("marker is not available, might be optional...")
14
+ logger.warning("marker is not available, might be optional...")
15
15
  MarkerPipeline = None
16
16
 
17
17
 
18
18
  try:
19
19
  from .miner_u import MinerUPipeline
20
20
  except ModuleNotFoundError:
21
- logger.exception("mineru is not available, might be optional...")
21
+ logger.warning("mineru is not available, might be optional...")
22
22
  MinerUPipeline = None
23
23
 
24
24
 
@@ -0,0 +1,9 @@
1
+ from .cli import cli_app
2
+
3
+
4
+ def main() -> None:
5
+ cli_app()
6
+
7
+
8
+ if __name__ == "__main__":
9
+ main()
@@ -0,0 +1,50 @@
1
+ import importlib
2
+ import logging
3
+ import os
4
+ from typing import Annotated
5
+
6
+ import typer
7
+ from icij_common.logging_utils import setup_loggers
8
+
9
+ import extract_python
10
+
11
+ from .doc import configs_app
12
+ from .utils import AsyncTyper
13
+
14
+ cli_app = AsyncTyper(
15
+ context_settings={"help_option_names": ["-h", "--help"]},
16
+ pretty_exceptions_enable=False,
17
+ )
18
+ cli_app.add_typer(configs_app)
19
+
20
+
21
+ def version_callback(value: bool) -> None: # noqa: FBT001
22
+ if value:
23
+ package_version = importlib.metadata.version(extract_python.__name__)
24
+ print(package_version)
25
+ raise typer.Exit()
26
+
27
+
28
+ def pretty_exc_callback(value: bool) -> None: # noqa: FBT001
29
+ if not value:
30
+ os.environ["TYPER_STANDARD_TRACEBACK"] = "1"
31
+
32
+
33
+ @cli_app.callback()
34
+ def main(
35
+ version: Annotated[ # noqa: ARG001
36
+ bool | None,
37
+ typer.Option("--version", callback=version_callback, is_eager=True),
38
+ ] = None,
39
+ *,
40
+ pretty_exceptions: Annotated[ # noqa: ARG001
41
+ bool,
42
+ typer.Option(
43
+ "--pretty-exceptions", callback=pretty_exc_callback, is_eager=True
44
+ ),
45
+ ] = False,
46
+ ) -> None:
47
+ setup_loggers(
48
+ ["__main__", extract_python.__name__, "docling", "mineru", "marker"],
49
+ level=logging.DEBUG,
50
+ )
@@ -0,0 +1,78 @@
1
+ import logging
2
+ import os
3
+ from pathlib import Path
4
+ from typing import Annotated, Never
5
+
6
+ import typer
7
+ from extract_core import (
8
+ Device,
9
+ InputDoc,
10
+ OutputFormat,
11
+ Pipeline,
12
+ PipelineConfig,
13
+ PipelineSize,
14
+ PipelineType,
15
+ )
16
+ from extract_core.default_config import default_config
17
+ from pydantic import TypeAdapter
18
+
19
+ from .utils import AsyncTyper
20
+
21
+ logger = logging.getLogger(__name__)
22
+
23
+ _DOC = "doc"
24
+
25
+ _DOC_CONVERT_HELP = "Converts documents to the provided output format"
26
+ _DOC_CONVERT_DOCS_HELP = "input documents (can be file or directories)"
27
+ _DOC_CONVERT_OUTPUT_PATH_HELP = "directory where the converted documents will be saved"
28
+ _DOC_CONVERT_DEVICE_HELP = "device used to run inference"
29
+ _DOC_CONVERT_OUTPUT_FMT_HELP = "output format"
30
+ _DOC_CONVERT_PIPELINE_CONFIG = "path to a JSON pipeline configuration"
31
+
32
+
33
+ configs_app = AsyncTyper(name=_DOC)
34
+
35
+
36
+ @configs_app.async_command(help=_DOC_CONVERT_HELP)
37
+ async def convert(
38
+ docs: Annotated[list[Path], typer.Argument(help=_DOC_CONVERT_DOCS_HELP)],
39
+ output_path: Annotated[Path, typer.Argument(help=_DOC_CONVERT_OUTPUT_PATH_HELP)],
40
+ output_format: Annotated[
41
+ OutputFormat, typer.Option(help=_DOC_CONVERT_OUTPUT_FMT_HELP)
42
+ ] = OutputFormat.MARKDOWN,
43
+ config: Annotated[
44
+ Path | None,
45
+ typer.Option("-c", "--config", help=_DOC_CONVERT_PIPELINE_CONFIG),
46
+ ] = None,
47
+ ) -> None:
48
+ if config is not None:
49
+ config = TypeAdapter(PipelineConfig).validate_json(config.read_text())
50
+ else:
51
+ config = default_config(
52
+ pipeline_type=PipelineType.DOCLING,
53
+ device=Device.CPU,
54
+ size=PipelineSize.SMALL,
55
+ )
56
+ config = config.to_device()
57
+ docs = [InputDoc.from_path(d, n_pages=1) for p in docs for d in _walkdir(p)]
58
+ pipeline = Pipeline.from_config(config)
59
+ output_paths = []
60
+ async for res in pipeline.extract_content(docs, output_format, output_path):
61
+ output_path = res.output.path
62
+ logger.info("converted %s to %s", res.input.path, output_path / output_path)
63
+ output_paths.append(output_path / output_path)
64
+ print("\n".join(output_paths))
65
+
66
+
67
+ def _raise(exc: OSError) -> Never:
68
+ raise exc
69
+
70
+
71
+ def _walkdir(path: Path) -> list[Path]:
72
+ if path.is_file():
73
+ return [path]
74
+ paths = []
75
+ for root, _, files in os.walk(path, onerror=_raise):
76
+ root = Path(root) # noqa: PLW2901
77
+ paths.extend(root / f for f in files)
78
+ return paths
@@ -0,0 +1,25 @@
1
+ import asyncio
2
+ import sys
3
+ from collections.abc import Callable
4
+ from functools import wraps
5
+ from typing import Any
6
+
7
+ import typer
8
+
9
+
10
+ class AsyncTyper(typer.Typer):
11
+ def async_command(self, *args, **kwargs) -> Callable[[Callable], Callable]:
12
+ def decorator(async_func: Callable) -> Callable:
13
+ @wraps(async_func)
14
+ def sync_func(*_args, **_kwargs) -> Any:
15
+ res = asyncio.run(async_func(*_args, **_kwargs))
16
+ return res
17
+
18
+ self.command(*args, **kwargs)(sync_func)
19
+ return async_func
20
+
21
+ return decorator
22
+
23
+
24
+ def eprint(*args, **kwargs) -> None:
25
+ print(*args, file=sys.stderr, **kwargs)
@@ -59,8 +59,7 @@ class DoclingPipeline(Pipeline):
59
59
  def __init__(self, config: DoclingPipelineConfig):
60
60
  super().__init__(config)
61
61
  format_options = {
62
- k: v.to_docling(self._device)
63
- for k, v in self._config.format_options.items()
62
+ k: v.to_docling() for k, v in self._config.format_options.items()
64
63
  }
65
64
  logger.info(
66
65
  "resolved format options to: %s",
@@ -14,6 +14,10 @@ dependencies = [
14
14
  ]
15
15
 
16
16
  [project.optional-dependencies]
17
+ cli = [
18
+ "typer~=0.21.2",
19
+ ]
20
+
17
21
  benches = [
18
22
  "html2image~=2.0.7",
19
23
  "markdown2>=2.5.4",
@@ -51,10 +55,16 @@ required-environments = [
51
55
  "sys_platform == 'darwin' and platform_machine == 'arm64'",
52
56
  "sys_platform == 'linux'",
53
57
  ]
58
+ exclude-dependencies = [
59
+ "mlx-vlm",
60
+ ]
54
61
  override-dependencies = [
55
62
  "pillow==11.3.0",
56
63
  ]
57
64
 
65
+ [project.scripts]
66
+ extract_python = "extract_python.__main__:main"
67
+
58
68
  [[tool.uv.index]]
59
69
  name = "pytorch-cpu"
60
70
  url = "https://download.pytorch.org/whl/cpu"
@@ -26,6 +26,7 @@ conflicts = [[
26
26
 
27
27
  [manifest]
28
28
  overrides = [{ name = "pillow", specifier = "==11.3.0" }]
29
+ excludes = ["mlx-vlm"]
29
30
 
30
31
  [[package]]
31
32
  name = "accelerate"
@@ -279,6 +280,32 @@ wheels = [
279
280
  { url = "https://files.pythonhosted.org/packages/f6/22/91616fe707a5c5510de2cac9b046a30defe7007ba8a0c04f9c08f27df312/audioop_lts-0.2.2-cp314-cp314t-win_arm64.whl", hash = "sha256:b492c3b040153e68b9fdaff5913305aaaba5bb433d8a7f73d5cf6a64ed3cc1dd", size = 25206, upload-time = "2025-08-05T16:43:16.444Z" },
280
281
  ]
281
282
 
283
+ [[package]]
284
+ name = "av"
285
+ version = "19.0.1"
286
+ source = { registry = "https://pypi.org/simple" }
287
+ sdist = { url = "https://files.pythonhosted.org/packages/90/bc/a2a40e503250fe5d4174471911828f31658864eb69a8a7cb960c715e17b7/av-19.0.1.tar.gz", hash = "sha256:08674930eaf1af78a3ed8f93d3ba49383323b3a867e84349d9c399e36f7497da", size = 4274648, upload-time = "2026-10-03T01:48:28.575Z" }
288
+ wheels = [
289
+ { url = "https://files.pythonhosted.org/packages/ec/2f/f4d219b2c72fea88bcbaea23de5b7f864ebecd348586fd2fe69f7f657147/av-19.0.1-cp312-abi3-macosx_11_0_x86_64.whl", hash = "sha256:2bd44ef4c09bb04aa6100d4c6191ddedaffef6af757ac55d5b4dc90915859299", size = 22625494, upload-time = "2026-10-03T01:47:21.866Z" },
290
+ { url = "https://files.pythonhosted.org/packages/ff/75/db37bb43a12a317cc0c0b96ddabc7896f582503b377e0803d4d721969522/av-19.0.1-cp312-abi3-macosx_14_0_arm64.whl", hash = "sha256:29d85e4ee36bf8f475dad07d4f4417c07bba62535f6a7179429c357e0ca8fb0f", size = 18439188, upload-time = "2026-10-03T01:47:25.541Z" },
291
+ { url = "https://files.pythonhosted.org/packages/10/4b/61f138fcf21e7bb50655ed21dd7fdc7a296baf72ea3c7ad8e89cb00b69c1/av-19.0.1-cp312-abi3-manylinux_2_28_aarch64.whl", hash = "sha256:437d4c0d5a7d771f2c3af84cd28e6aac6e173851116c60b53e81dbf1eebe4eab", size = 32676941, upload-time = "2026-10-03T01:47:29.237Z" },
292
+ { url = "https://files.pythonhosted.org/packages/c8/97/5fb45934ac64e8afc2c6869a7dcb8cb2af1ddab09a725367548856cbb59f/av-19.0.1-cp312-abi3-manylinux_2_28_x86_64.whl", hash = "sha256:1bea5b6134209305199bce7627ac3d33964de2cf2b09c77d08e7f67cf8bd4170", size = 34983451, upload-time = "2026-10-03T01:47:32.895Z" },
293
+ { url = "https://files.pythonhosted.org/packages/66/f2/6eee1b99ac492fa1965d6fd466ef8b644ca296b4f1dfa8c8225ab340b139/av-19.0.1-cp312-abi3-manylinux_2_31_armv7l.whl", hash = "sha256:1de938ec0134ad88f795dfe0a2dfc2d59e9ecea39a20158d37961279a3483612", size = 41660680, upload-time = "2026-10-03T01:47:36.903Z" },
294
+ { url = "https://files.pythonhosted.org/packages/11/be/e4ddd0197d02a3114402f3ffde541f6c4edecd24d670bea0da1eb6f15fb2/av-19.0.1-cp312-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:bcd0af218ecbeddbb1b0c56c4278043a3d97b87f3b8e33f6f92d452c744b1b08", size = 33748455, upload-time = "2026-10-03T01:47:40.541Z" },
295
+ { url = "https://files.pythonhosted.org/packages/7a/41/b9af863f635f64abaf5eb734521306487fc79447f5d55d792339a81c8a4d/av-19.0.1-cp312-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:935a6b6386a6994964e324eb02af4dab01eedbcbbde23b4b21bf1dc59b004244", size = 36008899, upload-time = "2026-10-03T01:47:44.13Z" },
296
+ { url = "https://files.pythonhosted.org/packages/e6/dc/a87a5a5e3ac462734f9befd8bad1447301e5802d8c111e22bf708fba7af3/av-19.0.1-cp312-abi3-win_amd64.whl", hash = "sha256:906fc3db09288319a75ea23ffefb59961c7dbe0d1c074601507a89de7d8593d8", size = 28149519, upload-time = "2026-10-03T01:47:47.372Z" },
297
+ { url = "https://files.pythonhosted.org/packages/a5/78/16864f1aa2c3ac5017f15132b85c6d3c74bb85caca8c45ce836ad30dfe20/av-19.0.1-cp312-abi3-win_arm64.whl", hash = "sha256:e9e1b0cae6cebd2adc2c5c6691fc890112f8f6c846b76a9135307617db1e32e9", size = 20706822, upload-time = "2026-10-03T01:47:50.72Z" },
298
+ { url = "https://files.pythonhosted.org/packages/78/4a/b5d7614856af72d7c18b926dda43bd227844b0b42d64e7c478b080f8d9c1/av-19.0.1-cp314-cp314t-macosx_11_0_x86_64.whl", hash = "sha256:3ef376ab828730f50b635e3541f305503adad713cb4c3eadb5ad0e4c6a6f4a72", size = 22909764, upload-time = "2026-10-03T01:47:54.032Z" },
299
+ { url = "https://files.pythonhosted.org/packages/b6/c9/50b2dedd4314a0ba0d78d7a7a52f7b073bc3377e5152e51d9d5627c5bcf4/av-19.0.1-cp314-cp314t-macosx_14_0_arm64.whl", hash = "sha256:17f2e42a1c969c78c616fe58bc69641a9df404c1ac2f01b50c1ddc22e5c31f69", size = 18718945, upload-time = "2026-10-03T01:47:58.396Z" },
300
+ { url = "https://files.pythonhosted.org/packages/ef/a5/eb2b6aadbda16ee676c76e43012709f0cdfe09c35bc9ad4ffb5099827e72/av-19.0.1-cp314-cp314t-manylinux_2_28_aarch64.whl", hash = "sha256:aafd294abd0e5c23e6c813b10fb4792cf1dd1002c1aead0292d195cda2ca154e", size = 36470355, upload-time = "2026-10-03T01:48:01.686Z" },
301
+ { url = "https://files.pythonhosted.org/packages/c1/f0/25e7d21cc29e949118bdac6efe0ef5c5020fc4273a3ea237989728ebe816/av-19.0.1-cp314-cp314t-manylinux_2_28_x86_64.whl", hash = "sha256:400ba5234865dc370c442658efff0672c64dcad2de26a2a7c900abf16ffd9f68", size = 38457564, upload-time = "2026-10-03T01:48:05.61Z" },
302
+ { url = "https://files.pythonhosted.org/packages/3f/09/77fec7c8de49fb815d55de1dfac21b39fb9e6915cbd8dcd945538ebb6f44/av-19.0.1-cp314-cp314t-manylinux_2_31_armv7l.whl", hash = "sha256:5e527b9d2d23c096d2b488e19a40ceba3654ea84a3cecee1c1b46c70ceaceae2", size = 43462245, upload-time = "2026-10-03T01:48:10.674Z" },
303
+ { url = "https://files.pythonhosted.org/packages/8c/1d/bb0281ada4203c5d85f7e8b045de2cadc89c3b5d0ed5705298f7a9288b1f/av-19.0.1-cp314-cp314t-musllinux_1_2_aarch64.whl", hash = "sha256:79136e62d4bc93db81fb63d6dd0060e86259426c071ca5157b1abe8c815c40b7", size = 37339005, upload-time = "2026-10-03T01:48:14.805Z" },
304
+ { url = "https://files.pythonhosted.org/packages/0a/84/19a9d37d7546a3879d759a8957b2513a029cafb81f60218c496b1ce9d5a8/av-19.0.1-cp314-cp314t-musllinux_1_2_x86_64.whl", hash = "sha256:330f91c704aa822b96d9aa21382c0eb41a68531d388078d724d334faa460cbcc", size = 39466754, upload-time = "2026-10-03T01:48:18.988Z" },
305
+ { url = "https://files.pythonhosted.org/packages/30/c4/39d4e2b778f1e86672671e25c3fd38e8d59d59b6f65c5cd13d7fae3d88a3/av-19.0.1-cp314-cp314t-win_amd64.whl", hash = "sha256:8289295bfd2a438f2cf83c3ab426964055e441f1500410a842e7a767bdc8e51e", size = 29063526, upload-time = "2026-10-03T01:48:22.724Z" },
306
+ { url = "https://files.pythonhosted.org/packages/f4/7d/a20ff44c1445c09a93985418f6997e5823635848e955a7953339636a9829/av-19.0.1-cp314-cp314t-win_arm64.whl", hash = "sha256:e1f70b1bda35588aff5fc526500376afe143e33cfce5d7e30d368170c38717db", size = 21915698, upload-time = "2026-10-03T01:48:26.386Z" },
307
+ ]
308
+
282
309
  [[package]]
283
310
  name = "babel"
284
311
  version = "2.18.0"
@@ -664,7 +691,7 @@ wheels = [
664
691
 
665
692
  [[package]]
666
693
  name = "docling-core"
667
- version = "2.98.0"
694
+ version = "2.100.0"
668
695
  source = { registry = "https://pypi.org/simple" }
669
696
  dependencies = [
670
697
  { name = "defusedxml" },
@@ -682,16 +709,17 @@ dependencies = [
682
709
  { name = "typer" },
683
710
  { name = "typing-extensions" },
684
711
  ]
685
- sdist = { url = "https://files.pythonhosted.org/packages/6f/1f/03528148da2a97c35fcf0a0a9bc7614b7585cf1f63f624fce85aafaa60c9/docling_core-2.98.0.tar.gz", hash = "sha256:a5dd76d747d23b6e1c83b434e29285798a2547d6e23a0ddfc487ed09ff5663ea", size = 383348, upload-time = "2026-09-22T09:42:51.273Z" }
712
+ sdist = { url = "https://files.pythonhosted.org/packages/4d/f7/5fe0635e3a3251d08eb4a30dffd02414b6614986431a671c5a4352fcf90f/docling_core-2.100.0.tar.gz", hash = "sha256:87a4f7318d774194364a3ac6c1e6b1df0517f28c7ac395d92aa3535eb3e5c446", size = 404111, upload-time = "2026-10-06T12:14:49.988Z" }
686
713
  wheels = [
687
- { url = "https://files.pythonhosted.org/packages/23/c6/479371d25ad9fd0137c3036bb21d0b1f96075b85958a81ce868aae4bc515/docling_core-2.98.0-py3-none-any.whl", hash = "sha256:655903b45f4c66c759842df8172237bda3ef35055197bf371fa5c855d3df0bd3", size = 305641, upload-time = "2026-09-22T09:42:49.591Z" },
714
+ { url = "https://files.pythonhosted.org/packages/19/4c/9d319f71922af600271cb6e36b15bae0969444182d55361c39b566e29ab7/docling_core-2.100.0-py3-none-any.whl", hash = "sha256:4cb57c4ccb5740e83e5e63d2215201711125e3603115bc720b099bcc49d09029", size = 316004, upload-time = "2026-10-06T12:14:48.161Z" },
688
715
  ]
689
716
 
690
717
  [package.optional-dependencies]
691
718
  chunking = [
692
719
  { name = "semchunk" },
693
720
  { name = "transformers", version = "4.57.6", source = { registry = "https://pypi.org/simple" }, marker = "extra == 'extra-14-extract-python-marker'" },
694
- { name = "transformers", version = "5.15.1", source = { registry = "https://pypi.org/simple" }, marker = "extra == 'extra-14-extract-python-mineru' or extra != 'extra-14-extract-python-marker'" },
721
+ { name = "transformers", version = "5.15.1", source = { registry = "https://pypi.org/simple" }, marker = "(sys_platform != 'darwin' and extra != 'extra-14-extract-python-marker') or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
722
+ { name = "transformers", version = "5.18.0", source = { registry = "https://pypi.org/simple" }, marker = "(sys_platform == 'darwin' and extra != 'extra-14-extract-python-marker') or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
695
723
  { name = "tree-sitter" },
696
724
  { name = "tree-sitter-c" },
697
725
  { name = "tree-sitter-javascript" },
@@ -714,7 +742,8 @@ dependencies = [
714
742
  { name = "torchvision", version = "0.27.0", source = { registry = "https://download.pytorch.org/whl/cpu" }, marker = "(platform_machine == 'arm64' and sys_platform == 'darwin') or (platform_machine != 'arm64' and extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru') or (sys_platform != 'darwin' and extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
715
743
  { name = "torchvision", version = "0.27.0+cpu", source = { registry = "https://download.pytorch.org/whl/cpu" }, marker = "platform_machine != 'arm64' or sys_platform != 'darwin' or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
716
744
  { name = "transformers", version = "4.57.6", source = { registry = "https://pypi.org/simple" }, marker = "extra == 'extra-14-extract-python-marker'" },
717
- { name = "transformers", version = "5.15.1", source = { registry = "https://pypi.org/simple" }, marker = "extra == 'extra-14-extract-python-mineru' or extra != 'extra-14-extract-python-marker'" },
745
+ { name = "transformers", version = "5.15.1", source = { registry = "https://pypi.org/simple" }, marker = "(sys_platform != 'darwin' and extra != 'extra-14-extract-python-marker') or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
746
+ { name = "transformers", version = "5.18.0", source = { registry = "https://pypi.org/simple" }, marker = "(sys_platform == 'darwin' and extra != 'extra-14-extract-python-marker') or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
718
747
  ]
719
748
  sdist = { url = "https://files.pythonhosted.org/packages/69/ee/7dd19487bf333320f0fc73eddcb42347d9f266310ebb1a605f120987f7c1/docling_ibm_models-4.0.3.tar.gz", hash = "sha256:509e89af75e06b48500977dca11ecb6022e81b1526c90dd54587a60ccc82bdb6", size = 77649, upload-time = "2026-09-18T16:15:36.731Z" }
720
749
  wheels = [
@@ -723,7 +752,7 @@ wheels = [
723
752
 
724
753
  [[package]]
725
754
  name = "docling-parse"
726
- version = "7.21.0"
755
+ version = "7.22.2"
727
756
  source = { registry = "https://pypi.org/simple" }
728
757
  dependencies = [
729
758
  { name = "docling-core" },
@@ -731,23 +760,23 @@ dependencies = [
731
760
  { name = "pydantic" },
732
761
  { name = "pywin32", marker = "sys_platform == 'win32' or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
733
762
  ]
734
- sdist = { url = "https://files.pythonhosted.org/packages/dc/d3/c0b03384b7735261662a5298837f5da49727ba0bf0d63baaeedc97236955/docling_parse-7.21.0.tar.gz", hash = "sha256:babbd8b43deb69706963ebc9b54283b458395a53519f68fd4bd0e7c5f793943b", size = 7035738, upload-time = "2026-09-22T09:53:15.861Z" }
763
+ sdist = { url = "https://files.pythonhosted.org/packages/bc/19/4036a3e844a6385b720b40f9c44d6ab4b6733707dd698dcfec3beda0a48d/docling_parse-7.22.2.tar.gz", hash = "sha256:368d076d2b887fcb44a90c37c219a190b975f01b257b61bd83ba75dc06861536", size = 7045499, upload-time = "2026-10-05T07:23:55.9Z" }
735
764
  wheels = [
736
- { url = "https://files.pythonhosted.org/packages/53/94/836d8fe9d079bc1d1d92498386e9ab7000bc7d1ae5154134e9677d8ec27a/docling_parse-7.21.0-cp313-cp313-macosx_14_0_arm64.whl", hash = "sha256:da68374d71ad50fde66318db0034e56159fb74bf2f73602084b986471f2faee6", size = 9878142, upload-time = "2026-09-22T09:52:54.478Z" },
737
- { url = "https://files.pythonhosted.org/packages/94/9e/dcd2704f99ac831bc736c0c2bf30cedeb168cba908536f825946bec9f729/docling_parse-7.21.0-cp313-cp313-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:3d554d9f12b5d8d3d3b239afdcfad53996b4912d6d8e61dfd7635ecb4b15663a", size = 10437326, upload-time = "2026-09-22T09:52:57.327Z" },
738
- { url = "https://files.pythonhosted.org/packages/1d/39/51f01a03c39e40939f0e8761bb399462ddc2cb9d95920f59f84615468d9e/docling_parse-7.21.0-cp313-cp313-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:40201d5ffbdae0d21d52ecfd5c5045167caec27330a66706c35abc128fd33f40", size = 10845161, upload-time = "2026-09-22T09:52:59.175Z" },
739
- { url = "https://files.pythonhosted.org/packages/91/f4/9b900d863146dcd42c9eb984aaa41b5bcaeffe51f8743ca13f9ae3f0ee1d/docling_parse-7.21.0-cp313-cp313-win_amd64.whl", hash = "sha256:10b612c08c537221d5e1ab927a2226602479801e4132d6ebeea131fee9a09c72", size = 11886411, upload-time = "2026-09-22T09:53:01.888Z" },
740
- { url = "https://files.pythonhosted.org/packages/bc/ff/9ed39e6bdc256df9a4f69b3e0c33dcd7b7cdc9a491ee5da6be9645fe4a47/docling_parse-7.21.0-cp313-cp313-win_arm64.whl", hash = "sha256:e3f8740bad16bbc1d264708153751787529a06777a0cb51e29242e5ad7f4fd30", size = 9160130, upload-time = "2026-09-22T09:53:04.334Z" },
741
- { url = "https://files.pythonhosted.org/packages/97/87/fa8ac21dddf36ee89531e0c8de316e50003885551a99ec0e6eccd76efb12/docling_parse-7.21.0-cp314-cp314-macosx_14_0_arm64.whl", hash = "sha256:fe2a63c0b92ec73fcc0db674ee57ed9972ea2661c08eaf40c099976acd58e708", size = 9878785, upload-time = "2026-09-22T09:53:06.226Z" },
742
- { url = "https://files.pythonhosted.org/packages/7b/f7/a29021e0953b4f562c673bb8adf9737a35fc34500680553a05cbfaf6c085/docling_parse-7.21.0-cp314-cp314-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:40bfdf4e7ede1557b58c1680aea1fba1c22bf512e786d2889973fe330178d8df", size = 10437964, upload-time = "2026-09-22T09:53:08.473Z" },
743
- { url = "https://files.pythonhosted.org/packages/43/1b/8058f7a4a96ea97ee4a6fe53525c3505b5ca8e95361dce82223c40595c58/docling_parse-7.21.0-cp314-cp314-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:8f79205a3ae29104720b6136ed581b0574b3f365984eea5239d9e89ba4196032", size = 10845654, upload-time = "2026-09-22T09:53:10.309Z" },
744
- { url = "https://files.pythonhosted.org/packages/48/f1/82049b4c466df0c87864481b80d874cf4c1880a4b7a70c587f608400aced/docling_parse-7.21.0-cp314-cp314-win_amd64.whl", hash = "sha256:d3c57efa2ab1759373c440a2d869d08ff1079a9a3244231c7167b1e85440de51", size = 12322863, upload-time = "2026-09-22T09:53:12.31Z" },
745
- { url = "https://files.pythonhosted.org/packages/f4/3f/a5d44d7f51a5f3e2e96620c59770de7a98f3299d98959c1ec9b289937f61/docling_parse-7.21.0-cp314-cp314-win_arm64.whl", hash = "sha256:8cda2b5a6e8e9933ef68a114f283dfa83f6a2876ae70468ed055dbb0cdc8914b", size = 9528726, upload-time = "2026-09-22T09:53:14.311Z" },
765
+ { url = "https://files.pythonhosted.org/packages/d2/61/05b5d2fbc5dcb3023aa980daaebfaa03dd2447e21271668cd6b3058552c2/docling_parse-7.22.2-cp313-cp313-macosx_14_0_arm64.whl", hash = "sha256:53156ac7c1e389857f23b118f8cd59331d3a215be94f9f9ce9ae3145e4125759", size = 9878343, upload-time = "2026-10-05T07:23:28.238Z" },
766
+ { url = "https://files.pythonhosted.org/packages/b9/d1/538efc18c43a0c6e678e6dea7c14d0599a14a0d6c8fc3b192bd6d6a92f6d/docling_parse-7.22.2-cp313-cp313-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:f8eb209efd837d7ba055bc582ae0aa978e32b6fa9851fb39632a1c303f2b0693", size = 10437835, upload-time = "2026-10-05T07:23:30.77Z" },
767
+ { url = "https://files.pythonhosted.org/packages/5a/3f/c004e1eef07e24059130e9fe97ac5870c11a187162f1f7eb93e9c20c3a75/docling_parse-7.22.2-cp313-cp313-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:73265a552c3f28bee591a8ca1e0dc38b51ee17f5051a626aef2fb625ee81f305", size = 10842974, upload-time = "2026-10-05T07:23:33.225Z" },
768
+ { url = "https://files.pythonhosted.org/packages/65/a0/8943af36a45726fe5b59c0ed213ef260ed0111309547d5b01b9820dcd5bf/docling_parse-7.22.2-cp313-cp313-win_amd64.whl", hash = "sha256:5a1433aceca0b1be834a79b6a7bddd89681426b8c1b06725aebe7740583d38a3", size = 11886875, upload-time = "2026-10-05T07:23:35.722Z" },
769
+ { url = "https://files.pythonhosted.org/packages/08/44/5d44f7ca3801fbcc5e0f6e51d92cb85a6952873203ecf56227503fb53f6a/docling_parse-7.22.2-cp313-cp313-win_arm64.whl", hash = "sha256:e23c55d856d0124634b42d8b63a879ecb1056769a3b38cb6ea186e03d20c0421", size = 9274980, upload-time = "2026-10-05T07:23:38.454Z" },
770
+ { url = "https://files.pythonhosted.org/packages/1f/71/75b2492ea7de1dea7b23405d8afbbf6be44e194857fad328c400dbcc9ef4/docling_parse-7.22.2-cp314-cp314-macosx_14_0_arm64.whl", hash = "sha256:9a003356d71578a56718abd343baf5c6d745ea7d7e988ca524b3af5bbd35760d", size = 9878449, upload-time = "2026-10-05T07:23:41.387Z" },
771
+ { url = "https://files.pythonhosted.org/packages/65/e4/82e930b4f534f77dc70871d3b5f7cc4672fa6b89320d14bcad95a6ae3a72/docling_parse-7.22.2-cp314-cp314-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:7699b216003ef5759ee4fa61bc7a403a6ce63afcbbf316cd17e4562d5b306e8b", size = 10439277, upload-time = "2026-10-05T07:23:44.53Z" },
772
+ { url = "https://files.pythonhosted.org/packages/f1/b2/6161ee6ab8e519177451620976abd1a002dbe1bdc98d4f8cddd7c49f68f4/docling_parse-7.22.2-cp314-cp314-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:a6a5aa4e0c0f62c4e0032c24117410d5773dbc303da5b4a282ff5abf9f491391", size = 10843921, upload-time = "2026-10-05T07:23:47.72Z" },
773
+ { url = "https://files.pythonhosted.org/packages/28/52/3593b031a4319379f9850dd3e8bc76adbcc62d7533f0741e6e3257d77d9c/docling_parse-7.22.2-cp314-cp314-win_amd64.whl", hash = "sha256:4bca9aeefa92896a12eb96e8858adc6b3adbd34c457095e5be68db84c6bb8c56", size = 12324184, upload-time = "2026-10-05T07:23:50.939Z" },
774
+ { url = "https://files.pythonhosted.org/packages/79/ec/c439cced7ed74407b0fd70d48e4de19b7a8e0da88f786b95bfb0af10c77f/docling_parse-7.22.2-cp314-cp314-win_arm64.whl", hash = "sha256:6bfff00d2577f288e037069bbf5087bcf01305f6b78b9432d599c6ffe03557a2", size = 9652594, upload-time = "2026-10-05T07:23:53.627Z" },
746
775
  ]
747
776
 
748
777
  [[package]]
749
778
  name = "docling-slim"
750
- version = "2.130.0"
779
+ version = "2.134.0"
751
780
  source = { registry = "https://pypi.org/simple" }
752
781
  dependencies = [
753
782
  { name = "certifi" },
@@ -760,9 +789,9 @@ dependencies = [
760
789
  { name = "requests" },
761
790
  { name = "tqdm" },
762
791
  ]
763
- sdist = { url = "https://files.pythonhosted.org/packages/0e/d5/4db4fd7e756d40db15b07ad072294548b0f508fb35fdd3496fa191223822/docling_slim-2.130.0.tar.gz", hash = "sha256:d45b467d322cd15f03fcc1e792bd4a52aa2dc075dfee229b13ae43a8603f82a6", size = 708768, upload-time = "2026-09-22T15:36:53.243Z" }
792
+ sdist = { url = "https://files.pythonhosted.org/packages/7b/75/ee898b4b7a310c187d0d7ad89002b000efeead0af177f9e50f71554b194b/docling_slim-2.134.0.tar.gz", hash = "sha256:63f12c79bdd0bbe205f244f97f0a8640c590ea46de4d913d23e963ec57e531dd", size = 795345, upload-time = "2026-10-06T04:57:33.257Z" }
764
793
  wheels = [
765
- { url = "https://files.pythonhosted.org/packages/db/57/9e21add9ec842b314d4d193db8c4186354b872529ef3a17f8aeedbd0c515/docling_slim-2.130.0-py3-none-any.whl", hash = "sha256:913043ee069eb4f107d47ebfca8bdbfbfd7ec84160a7193b7260777e3adbea19", size = 876695, upload-time = "2026-09-22T15:36:51.06Z" },
794
+ { url = "https://files.pythonhosted.org/packages/cd/58/526bc8f7c7f46dc22afa6f548c67e0bd685cf5f97faf7ee3d96c26355732/docling_slim-2.134.0-py3-none-any.whl", hash = "sha256:818b350e1e9878c0d54bd4ff972bd1918cc9827f193257162cc6f3a797175c29", size = 974718, upload-time = "2026-10-06T04:57:31.022Z" },
766
795
  ]
767
796
 
768
797
  [package.optional-dependencies]
@@ -773,10 +802,28 @@ feat-ocr-easyocr = [
773
802
  feat-ocr-mac = [
774
803
  { name = "ocrmac", marker = "sys_platform == 'darwin' or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
775
804
  ]
805
+ feat-ocr-rapidocr = [
806
+ { name = "rapidocr" },
807
+ ]
776
808
  feat-ocr-tesserocr = [
777
809
  { name = "pandas" },
778
810
  { name = "tesserocr" },
779
811
  ]
812
+ models-onnxruntime = [
813
+ { name = "onnxruntime", marker = "sys_platform == 'darwin' or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
814
+ { name = "onnxruntime-gpu", marker = "sys_platform == 'linux' or sys_platform == 'win32' or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
815
+ ]
816
+ models-vlm-inline = [
817
+ { name = "accelerate" },
818
+ { name = "einops" },
819
+ { name = "open-clip-torch" },
820
+ { name = "peft" },
821
+ { name = "qwen-vl-utils" },
822
+ { name = "timm" },
823
+ { name = "transformers", version = "4.57.6", source = { registry = "https://pypi.org/simple" }, marker = "extra == 'extra-14-extract-python-marker'" },
824
+ { name = "transformers", version = "5.15.1", source = { registry = "https://pypi.org/simple" }, marker = "(sys_platform != 'darwin' and extra != 'extra-14-extract-python-marker') or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
825
+ { name = "transformers", version = "5.18.0", source = { registry = "https://pypi.org/simple" }, marker = "(sys_platform == 'darwin' and extra != 'extra-14-extract-python-marker') or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
826
+ ]
780
827
  standard = [
781
828
  { name = "accelerate" },
782
829
  { name = "beautifulsoup4" },
@@ -917,7 +964,7 @@ dependencies = [
917
964
 
918
965
  [package.metadata]
919
966
  requires-dist = [
920
- { name = "docling-slim", specifier = "~=2.127" },
967
+ { name = "docling-slim", specifier = "~=2.134" },
921
968
  { name = "icij-common", specifier = "~=0.8.2" },
922
969
  { name = "pydantic", specifier = "~=2.13" },
923
970
  { name = "pydantic-extra-types", extras = ["pycountry"], specifier = "~=2.11.1" },
@@ -956,8 +1003,14 @@ benches = [
956
1003
  { name = "pypdfium2", version = "4.30.0", source = { registry = "https://pypi.org/simple" }, marker = "extra == 'extra-14-extract-python-marker'" },
957
1004
  { name = "pypdfium2", version = "5.13.0", source = { registry = "https://pypi.org/simple" }, marker = "extra == 'extra-14-extract-python-mineru' or extra != 'extra-14-extract-python-marker'" },
958
1005
  ]
1006
+ cli = [
1007
+ { name = "typer" },
1008
+ ]
959
1009
  docling = [
960
- { name = "docling-slim", extra = ["feat-ocr-easyocr", "feat-ocr-mac", "feat-ocr-tesserocr", "standard"] },
1010
+ { name = "docling-slim", extra = ["feat-ocr-easyocr", "feat-ocr-mac", "feat-ocr-rapidocr", "feat-ocr-tesserocr", "standard"] },
1011
+ ]
1012
+ docling-gpu = [
1013
+ { name = "docling-slim", extra = ["models-onnxruntime", "models-vlm-inline"] },
961
1014
  ]
962
1015
  marker = [
963
1016
  { name = "marker-pdf" },
@@ -982,7 +1035,8 @@ dev = [
982
1035
 
983
1036
  [package.metadata]
984
1037
  requires-dist = [
985
- { name = "docling-slim", extras = ["feat-ocr-easyocr", "feat-ocr-mac", "feat-ocr-tesserocr", "standard"], marker = "extra == 'docling'", specifier = "~=2.127" },
1038
+ { name = "docling-slim", extras = ["feat-ocr-easyocr", "feat-ocr-mac", "feat-ocr-rapidocr", "feat-ocr-tesserocr", "standard"], marker = "extra == 'docling'", specifier = "~=2.127" },
1039
+ { name = "docling-slim", extras = ["feat-ocr-nemotron", "models-onnxruntime", "models-vlm-inline"], marker = "extra == 'docling-gpu'", specifier = "~=2.127" },
986
1040
  { name = "extract-core", editable = "../extract-core" },
987
1041
  { name = "html2image", marker = "extra == 'benches'", specifier = "~=2.0.7" },
988
1042
  { name = "icij-common", specifier = "~=0.8.2" },
@@ -995,8 +1049,9 @@ requires-dist = [
995
1049
  { name = "pypdfium2", marker = "extra == 'benches'", specifier = ">=4.30.0" },
996
1050
  { name = "python-pptx", marker = "extra == 'mineru'", specifier = "~=1.0" },
997
1051
  { name = "six", marker = "extra == 'mineru'", specifier = "~=1.17" },
1052
+ { name = "typer", marker = "extra == 'cli'", specifier = "~=0.21.2" },
998
1053
  ]
999
- provides-extras = ["benches", "docling", "marker", "mineru"]
1054
+ provides-extras = ["benches", "cli", "docling", "docling-gpu", "marker", "mineru"]
1000
1055
 
1001
1056
  [package.metadata.requires-dev]
1002
1057
  dev = [
@@ -2227,11 +2282,12 @@ dependencies = [
2227
2282
  { name = "safetensors", marker = "platform_machine == 'arm64' and sys_platform == 'darwin'" },
2228
2283
  { name = "shapely" },
2229
2284
  { name = "tinycss2" },
2230
- { name = "tokenizers" },
2285
+ { name = "tokenizers", version = "0.22.2", source = { registry = "https://pypi.org/simple" }, marker = "(sys_platform != 'darwin' and extra == 'extra-14-extract-python-mineru') or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
2286
+ { name = "tokenizers", version = "0.23.2", source = { registry = "https://pypi.org/simple" }, marker = "(sys_platform == 'darwin' and extra == 'extra-14-extract-python-mineru') or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
2231
2287
  { name = "torch", version = "2.12.0", source = { registry = "https://download.pytorch.org/whl/cpu" }, marker = "platform_machine == 'arm64' and sys_platform == 'darwin'" },
2232
2288
  { name = "torchvision", version = "0.27.0", source = { registry = "https://download.pytorch.org/whl/cpu" }, marker = "platform_machine == 'arm64' and sys_platform == 'darwin'" },
2233
2289
  { name = "tqdm" },
2234
- { name = "transformers", version = "5.15.1", source = { registry = "https://pypi.org/simple" }, marker = "platform_machine == 'arm64' and sys_platform == 'darwin'" },
2290
+ { name = "transformers", version = "5.18.0", source = { registry = "https://pypi.org/simple" }, marker = "platform_machine == 'arm64' and sys_platform == 'darwin'" },
2235
2291
  { name = "typer" },
2236
2292
  { name = "typing-extensions" },
2237
2293
  { name = "uvicorn" },
@@ -2249,7 +2305,8 @@ torch = [
2249
2305
  { name = "torch", version = "2.12.0+cpu", source = { registry = "https://download.pytorch.org/whl/cpu" }, marker = "(platform_machine != 'arm64' and extra == 'extra-14-extract-python-mineru') or (sys_platform != 'darwin' and extra == 'extra-14-extract-python-mineru') or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
2250
2306
  { name = "torchvision", version = "0.27.0", source = { registry = "https://download.pytorch.org/whl/cpu" }, marker = "(platform_machine == 'arm64' and sys_platform == 'darwin' and extra == 'extra-14-extract-python-mineru') or (platform_machine != 'arm64' and extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru') or (sys_platform != 'darwin' and extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
2251
2307
  { name = "torchvision", version = "0.27.0+cpu", source = { registry = "https://download.pytorch.org/whl/cpu" }, marker = "(platform_machine != 'arm64' and extra == 'extra-14-extract-python-mineru') or (sys_platform != 'darwin' and extra == 'extra-14-extract-python-mineru') or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
2252
- { name = "transformers", version = "5.15.1", source = { registry = "https://pypi.org/simple" } },
2308
+ { name = "transformers", version = "5.15.1", source = { registry = "https://pypi.org/simple" }, marker = "(sys_platform != 'darwin' and extra == 'extra-14-extract-python-mineru') or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
2309
+ { name = "transformers", version = "5.18.0", source = { registry = "https://pypi.org/simple" }, marker = "(sys_platform == 'darwin' and extra == 'extra-14-extract-python-mineru') or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
2253
2310
  ]
2254
2311
 
2255
2312
  [[package]]
@@ -2624,10 +2681,10 @@ name = "onnxruntime"
2624
2681
  version = "1.27.0"
2625
2682
  source = { registry = "https://pypi.org/simple" }
2626
2683
  dependencies = [
2627
- { name = "flatbuffers" },
2628
- { name = "numpy" },
2629
- { name = "packaging" },
2630
- { name = "protobuf" },
2684
+ { name = "flatbuffers", marker = "sys_platform == 'darwin' or extra == 'extra-14-extract-python-mineru'" },
2685
+ { name = "numpy", marker = "sys_platform == 'darwin' or extra == 'extra-14-extract-python-mineru'" },
2686
+ { name = "packaging", marker = "sys_platform == 'darwin' or extra == 'extra-14-extract-python-mineru'" },
2687
+ { name = "protobuf", marker = "sys_platform == 'darwin' or extra == 'extra-14-extract-python-mineru'" },
2631
2688
  ]
2632
2689
  wheels = [
2633
2690
  { url = "https://files.pythonhosted.org/packages/fb/2b/54208fd03ad410480bc17edf4869376362da8bbf46fe186ddf4cb5cc20fe/onnxruntime-1.27.0-cp313-cp313-macosx_14_0_arm64.whl", hash = "sha256:b3e5b58b8c89c2b20e086e890aa9527377e5c240dc3ecc1640d18e07705eeb1c", size = 18432958, upload-time = "2026-06-15T22:42:53.105Z" },
@@ -2646,6 +2703,52 @@ wheels = [
2646
2703
  { url = "https://files.pythonhosted.org/packages/b7/f6/2bac21f722aa45d876d4a51f26bd0ef30e704068a3cd5021a5a7cd784271/onnxruntime-1.27.0-cp314-cp314t-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:370d211e1ceeac4cd5f45301655463ac59e27cdc74d9f7aeb2d19ff4b7a76715", size = 18670781, upload-time = "2026-06-15T22:43:17.151Z" },
2647
2704
  ]
2648
2705
 
2706
+ [[package]]
2707
+ name = "onnxruntime-gpu"
2708
+ version = "1.30.0"
2709
+ source = { registry = "https://pypi.org/simple" }
2710
+ dependencies = [
2711
+ { name = "flatbuffers", marker = "(sys_platform != 'darwin' and sys_platform != 'emscripten') or (sys_platform == 'darwin' and extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru') or (sys_platform == 'emscripten' and extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
2712
+ { name = "numpy", marker = "(sys_platform != 'darwin' and sys_platform != 'emscripten') or (sys_platform == 'darwin' and extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru') or (sys_platform == 'emscripten' and extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
2713
+ { name = "packaging", marker = "(sys_platform != 'darwin' and sys_platform != 'emscripten') or (sys_platform == 'darwin' and extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru') or (sys_platform == 'emscripten' and extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
2714
+ { name = "protobuf", marker = "(sys_platform != 'darwin' and sys_platform != 'emscripten') or (sys_platform == 'darwin' and extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru') or (sys_platform == 'emscripten' and extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
2715
+ ]
2716
+ wheels = [
2717
+ { url = "https://files.pythonhosted.org/packages/31/ae/cfec0c21d039e4125fdc333e2e2032ed0a49eb50f9dd70b7f3819b339018/onnxruntime_gpu-1.30.0-cp313-cp313-manylinux_2_28_x86_64.whl", hash = "sha256:9aba23f903f434cc55d851aeb801afee19e287c202f4ab8f59d02f04caa90aed", size = 246688669, upload-time = "2026-09-10T16:13:21.504Z" },
2718
+ { url = "https://files.pythonhosted.org/packages/07/cd/99171c84362c9a3292ec909c2e8edd922dec21d65e28e83e9d809ad178c6/onnxruntime_gpu-1.30.0-cp313-cp313-manylinux_2_34_aarch64.whl", hash = "sha256:0c68878729527a28202af6c1b7ebc2fc5a687e9f6e02bfb8796e39d1fe775fd7", size = 205732148, upload-time = "2026-09-10T15:53:16.111Z" },
2719
+ { url = "https://files.pythonhosted.org/packages/7e/d3/50fba71f4174f25f1e677ffe4b2ebc9f6c2a2aa8093c12e82c93792fd4e8/onnxruntime_gpu-1.30.0-cp313-cp313-win_amd64.whl", hash = "sha256:cf494da233f3fc02bbeb2a3178e3ace7361b6b4fef7031b19c1e5aed5e9328ce", size = 160479611, upload-time = "2026-09-10T15:56:28.799Z" },
2720
+ { url = "https://files.pythonhosted.org/packages/8e/41/3ebd7b9648259adcf3abd245be862a0cb4c24d4b12f4f7b82d3e9fcf18f0/onnxruntime_gpu-1.30.0-cp313-cp313t-manylinux_2_28_x86_64.whl", hash = "sha256:675d519cab613bc42685ecb998f8e8e8c6d73143496681650d30ef14fedeade7", size = 246684982, upload-time = "2026-09-10T16:13:31.799Z" },
2721
+ { url = "https://files.pythonhosted.org/packages/f5/b0/6187754ea45ce90c116730115e4d856a6107b9e708580093c112def9a567/onnxruntime_gpu-1.30.0-cp313-cp313t-manylinux_2_34_aarch64.whl", hash = "sha256:c6df4b62538f98482dd5cebc91cba668c4b82156e872517c314a383c904a191d", size = 205776779, upload-time = "2026-09-10T15:53:24.926Z" },
2722
+ { url = "https://files.pythonhosted.org/packages/bb/75/09a6c97136c747608867a63114d988384e6ddb5507e08bbe9ced08744a4c/onnxruntime_gpu-1.30.0-cp314-cp314-manylinux_2_28_x86_64.whl", hash = "sha256:5aacc98f394a13f4891c0a07e60479d8fb0b1ea6f3328b6b38cfc88f9dfb05ea", size = 246693192, upload-time = "2026-09-10T16:13:41.775Z" },
2723
+ { url = "https://files.pythonhosted.org/packages/73/93/415e8100a3d7d4754f406fadbe86fa91d03058a9a6d7fa8a692722e9ceb8/onnxruntime_gpu-1.30.0-cp314-cp314-manylinux_2_34_aarch64.whl", hash = "sha256:96cb15cff2f0dffaef3ae636560145d41be47bb73e3230113f2e946ca576014b", size = 205736852, upload-time = "2026-09-10T15:53:34.204Z" },
2724
+ { url = "https://files.pythonhosted.org/packages/50/fe/ed63f743d546fd089f98cfd415790b231ae8d3dba2208df8d22811f788a5/onnxruntime_gpu-1.30.0-cp314-cp314-win_amd64.whl", hash = "sha256:794fc118cdbf340ed02b44c7527e557ea51746002d48105d493ea41f0c9a60d0", size = 161124661, upload-time = "2026-09-10T15:56:36.489Z" },
2725
+ { url = "https://files.pythonhosted.org/packages/75/66/11c43d3d544380271aa14fd55c1e5be65f28fac007128e369f1192c9bfa9/onnxruntime_gpu-1.30.0-cp314-cp314t-manylinux_2_28_x86_64.whl", hash = "sha256:a4dfbe80d4167ff5bc269f266778ee6d0eb771fd9f3c0c4d83757e7ae42fe3c8", size = 246687044, upload-time = "2026-09-10T16:13:52.095Z" },
2726
+ { url = "https://files.pythonhosted.org/packages/74/3d/33878b36bc473415ff7ed92ba751a5b5181d2779c74441c3584c68261dbc/onnxruntime_gpu-1.30.0-cp314-cp314t-manylinux_2_34_aarch64.whl", hash = "sha256:13d4cf97120882739da8572ba26e2a0f39caed4c69749a147d8f1a77f74a4690", size = 205750897, upload-time = "2026-09-10T15:53:43.295Z" },
2727
+ ]
2728
+
2729
+ [[package]]
2730
+ name = "open-clip-torch"
2731
+ version = "3.3.0"
2732
+ source = { registry = "https://pypi.org/simple" }
2733
+ dependencies = [
2734
+ { name = "ftfy" },
2735
+ { name = "huggingface-hub", version = "0.36.2", source = { registry = "https://pypi.org/simple" }, marker = "extra == 'extra-14-extract-python-marker'" },
2736
+ { name = "huggingface-hub", version = "1.33.0", source = { registry = "https://pypi.org/simple" }, marker = "extra == 'extra-14-extract-python-mineru' or extra != 'extra-14-extract-python-marker'" },
2737
+ { name = "regex", version = "2024.11.6", source = { registry = "https://pypi.org/simple" }, marker = "extra == 'extra-14-extract-python-marker'" },
2738
+ { name = "regex", version = "2026.9.29", source = { registry = "https://pypi.org/simple" }, marker = "extra == 'extra-14-extract-python-mineru' or extra != 'extra-14-extract-python-marker'" },
2739
+ { name = "safetensors" },
2740
+ { name = "timm" },
2741
+ { name = "torch", version = "2.12.0", source = { registry = "https://download.pytorch.org/whl/cpu" }, marker = "(platform_machine == 'arm64' and sys_platform == 'darwin') or (platform_machine != 'arm64' and extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru') or (sys_platform != 'darwin' and extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
2742
+ { name = "torch", version = "2.12.0+cpu", source = { registry = "https://download.pytorch.org/whl/cpu" }, marker = "platform_machine != 'arm64' or sys_platform != 'darwin' or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
2743
+ { name = "torchvision", version = "0.27.0", source = { registry = "https://download.pytorch.org/whl/cpu" }, marker = "(platform_machine == 'arm64' and sys_platform == 'darwin') or (platform_machine != 'arm64' and extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru') or (sys_platform != 'darwin' and extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
2744
+ { name = "torchvision", version = "0.27.0+cpu", source = { registry = "https://download.pytorch.org/whl/cpu" }, marker = "platform_machine != 'arm64' or sys_platform != 'darwin' or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
2745
+ { name = "tqdm" },
2746
+ ]
2747
+ sdist = { url = "https://files.pythonhosted.org/packages/4a/1f/2bc9795047fa2c1ad2567ef78ce6dfc9a7b763fa534acee09a94da2a5b8f/open_clip_torch-3.3.0.tar.gz", hash = "sha256:904b1a9f909df8281bb3de60ab95491cd2994a509177ea4f9d6292a84fe24d6d", size = 1503380, upload-time = "2026-02-27T00:32:46.74Z" }
2748
+ wheels = [
2749
+ { url = "https://files.pythonhosted.org/packages/37/b5/41c315ccd94ca332ead3e832e83eee343ab245005c3e43d9d3e75eae34eb/open_clip_torch-3.3.0-py3-none-any.whl", hash = "sha256:c549ad5ed6bfc119cc11105033c0a2b9d7a2a4afeb40a58a09aab3da1a0043ce", size = 1547268, upload-time = "2026-02-27T00:32:44.902Z" },
2750
+ ]
2751
+
2649
2752
  [[package]]
2650
2753
  name = "openai"
2651
2754
  version = "1.109.1"
@@ -2828,6 +2931,31 @@ wheels = [
2828
2931
  { url = "https://files.pythonhosted.org/packages/bc/b9/4437bb89f04e57f48c96492a50d6168da5e201940de6620730d390449991/pdftext-0.6.3-py3-none-any.whl", hash = "sha256:528431ed8bdce39d74372cd3d27e8544af812f1f1adc81db229cf9fb48dacacb", size = 23693, upload-time = "2025-06-11T14:42:08.157Z" },
2829
2932
  ]
2830
2933
 
2934
+ [[package]]
2935
+ name = "peft"
2936
+ version = "0.21.2"
2937
+ source = { registry = "https://pypi.org/simple" }
2938
+ dependencies = [
2939
+ { name = "accelerate" },
2940
+ { name = "huggingface-hub", version = "0.36.2", source = { registry = "https://pypi.org/simple" }, marker = "extra == 'extra-14-extract-python-marker'" },
2941
+ { name = "huggingface-hub", version = "1.33.0", source = { registry = "https://pypi.org/simple" }, marker = "extra == 'extra-14-extract-python-mineru' or extra != 'extra-14-extract-python-marker'" },
2942
+ { name = "numpy" },
2943
+ { name = "packaging" },
2944
+ { name = "psutil" },
2945
+ { name = "pyyaml" },
2946
+ { name = "safetensors" },
2947
+ { name = "torch", version = "2.12.0", source = { registry = "https://download.pytorch.org/whl/cpu" }, marker = "(platform_machine == 'arm64' and sys_platform == 'darwin') or (platform_machine != 'arm64' and extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru') or (sys_platform != 'darwin' and extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
2948
+ { name = "torch", version = "2.12.0+cpu", source = { registry = "https://download.pytorch.org/whl/cpu" }, marker = "platform_machine != 'arm64' or sys_platform != 'darwin' or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
2949
+ { name = "tqdm" },
2950
+ { name = "transformers", version = "4.57.6", source = { registry = "https://pypi.org/simple" }, marker = "extra == 'extra-14-extract-python-marker'" },
2951
+ { name = "transformers", version = "5.15.1", source = { registry = "https://pypi.org/simple" }, marker = "(sys_platform != 'darwin' and extra != 'extra-14-extract-python-marker') or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
2952
+ { name = "transformers", version = "5.18.0", source = { registry = "https://pypi.org/simple" }, marker = "(sys_platform == 'darwin' and extra != 'extra-14-extract-python-marker') or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
2953
+ ]
2954
+ sdist = { url = "https://files.pythonhosted.org/packages/80/af/2e08abf1cd3b8792a02f5116808f2398c2f77ecdff801ced2ce16007a6f9/peft-0.21.2.tar.gz", hash = "sha256:b803ccfb3f3f316004d850284306687833a2235ea278fb56abc856203456142e", size = 983614, upload-time = "2026-10-01T10:26:36.129Z" }
2955
+ wheels = [
2956
+ { url = "https://files.pythonhosted.org/packages/70/0b/59441cdbfdd342ed03c08af90a2fe16173f0cd48ab219f523b39ca059a79/peft-0.21.2-py3-none-any.whl", hash = "sha256:106ab6077ff72c54d21577f9af5970e34bac7582cb14209f7b2b511e322a4eae", size = 835427, upload-time = "2026-10-01T10:26:34.085Z" },
2957
+ ]
2958
+
2831
2959
  [[package]]
2832
2960
  name = "pexpect"
2833
2961
  version = "4.9.0"
@@ -3706,6 +3834,21 @@ wheels = [
3706
3834
  { url = "https://files.pythonhosted.org/packages/81/d6/4bfbb40c9a0b42fc53c7cf442f6385db70b40f74a783130c5d0a5aa62228/pyzmq-27.1.0-cp314-cp314t-win_arm64.whl", hash = "sha256:dc5dbf68a7857b59473f7df42650c621d7e8923fb03fa74a526890f4d33cc4d7", size = 575170, upload-time = "2025-09-08T23:09:01.418Z" },
3707
3835
  ]
3708
3836
 
3837
+ [[package]]
3838
+ name = "qwen-vl-utils"
3839
+ version = "0.0.14"
3840
+ source = { registry = "https://pypi.org/simple" }
3841
+ dependencies = [
3842
+ { name = "av" },
3843
+ { name = "packaging" },
3844
+ { name = "pillow" },
3845
+ { name = "requests" },
3846
+ ]
3847
+ sdist = { url = "https://files.pythonhosted.org/packages/b6/b1/ad4fc2260a3badd278b38d642f3b987412f1f6682f0ef2b31b0572d5caa8/qwen_vl_utils-0.0.14.tar.gz", hash = "sha256:9c7cad5ae803b3a10f8bb7194deb12aeacdd032f92f4224e880c73587a7346ad", size = 8453, upload-time = "2025-09-23T09:38:57.532Z" }
3848
+ wheels = [
3849
+ { url = "https://files.pythonhosted.org/packages/c4/43/80f67e0336cb2fc725f8e06f7fe35c1d0fe946f4d2b8b2175e797e07349e/qwen_vl_utils-0.0.14-py3-none-any.whl", hash = "sha256:5e28657bfd031e56bd447c5901b58ddfc3835285ed100f4c56580e0ade054e96", size = 8120, upload-time = "2025-09-23T09:38:56.297Z" },
3850
+ ]
3851
+
3709
3852
  [[package]]
3710
3853
  name = "rapidfuzz"
3711
3854
  version = "3.14.5"
@@ -4592,6 +4735,25 @@ wheels = [
4592
4735
  { url = "https://files.pythonhosted.org/packages/81/59/208f71d70ddc6184f79b8c6d87d46eb7d7b12c19186a817dec9c9c3f3693/tifffile-2026.6.1-py3-none-any.whl", hash = "sha256:0d7382d2769b855b81ce358528e2b40c16d48aa39031746efa81215205332a8d", size = 267108, upload-time = "2026-05-31T23:57:10.597Z" },
4593
4736
  ]
4594
4737
 
4738
+ [[package]]
4739
+ name = "timm"
4740
+ version = "1.0.30"
4741
+ source = { registry = "https://pypi.org/simple" }
4742
+ dependencies = [
4743
+ { name = "huggingface-hub", version = "0.36.2", source = { registry = "https://pypi.org/simple" }, marker = "extra == 'extra-14-extract-python-marker'" },
4744
+ { name = "huggingface-hub", version = "1.33.0", source = { registry = "https://pypi.org/simple" }, marker = "extra == 'extra-14-extract-python-mineru' or extra != 'extra-14-extract-python-marker'" },
4745
+ { name = "pyyaml" },
4746
+ { name = "safetensors" },
4747
+ { name = "torch", version = "2.12.0", source = { registry = "https://download.pytorch.org/whl/cpu" }, marker = "(platform_machine == 'arm64' and sys_platform == 'darwin') or (platform_machine != 'arm64' and extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru') or (sys_platform != 'darwin' and extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
4748
+ { name = "torch", version = "2.12.0+cpu", source = { registry = "https://download.pytorch.org/whl/cpu" }, marker = "platform_machine != 'arm64' or sys_platform != 'darwin' or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
4749
+ { name = "torchvision", version = "0.27.0", source = { registry = "https://download.pytorch.org/whl/cpu" }, marker = "(platform_machine == 'arm64' and sys_platform == 'darwin') or (platform_machine != 'arm64' and extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru') or (sys_platform != 'darwin' and extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
4750
+ { name = "torchvision", version = "0.27.0+cpu", source = { registry = "https://download.pytorch.org/whl/cpu" }, marker = "platform_machine != 'arm64' or sys_platform != 'darwin' or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
4751
+ ]
4752
+ sdist = { url = "https://files.pythonhosted.org/packages/0e/ab/49041213990e48edcb6013463b18ef86d8d3c5d6ac5b408802c480443025/timm-1.0.30.tar.gz", hash = "sha256:1444a3af7e17fdba5440ac2bcf6e4b728da96dd62289ed9e1bcba31a545f5935", size = 2534738, upload-time = "2026-09-22T20:46:13.9Z" }
4753
+ wheels = [
4754
+ { url = "https://files.pythonhosted.org/packages/f5/85/b897716d0fa016af23b11e6526f5ef76d2013e8bc36b9c7ce15d9200e456/timm-1.0.30-py3-none-any.whl", hash = "sha256:c8e27adf6801a2fdfb570dbbdd171f2f084756eefc673df8cfaab06bd14864f3", size = 2668917, upload-time = "2026-09-22T20:46:12.11Z" },
4755
+ ]
4756
+
4595
4757
  [[package]]
4596
4758
  name = "tinycss2"
4597
4759
  version = "1.5.1"
@@ -4608,9 +4770,23 @@ wheels = [
4608
4770
  name = "tokenizers"
4609
4771
  version = "0.22.2"
4610
4772
  source = { registry = "https://pypi.org/simple" }
4773
+ resolution-markers = [
4774
+ "python_full_version >= '3.14' and sys_platform == 'win32'",
4775
+ "python_full_version >= '3.14' and sys_platform == 'emscripten'",
4776
+ "python_full_version >= '3.14' and platform_machine != 'arm64' and sys_platform == 'darwin'",
4777
+ "python_full_version >= '3.14' and platform_machine == 'aarch64' and sys_platform == 'linux'",
4778
+ "(python_full_version >= '3.14' and platform_machine != 'aarch64' and sys_platform == 'linux') or (python_full_version >= '3.14' and sys_platform != 'darwin' and sys_platform != 'emscripten' and sys_platform != 'linux' and sys_platform != 'win32')",
4779
+ "python_full_version < '3.14' and sys_platform == 'win32'",
4780
+ "python_full_version < '3.14' and sys_platform == 'emscripten'",
4781
+ "python_full_version < '3.14' and platform_machine != 'arm64' and sys_platform == 'darwin'",
4782
+ "python_full_version < '3.14' and platform_machine == 'aarch64' and sys_platform == 'linux'",
4783
+ "(python_full_version < '3.14' and platform_machine != 'aarch64' and sys_platform == 'linux') or (python_full_version < '3.14' and sys_platform != 'darwin' and sys_platform != 'emscripten' and sys_platform != 'linux' and sys_platform != 'win32')",
4784
+ "python_full_version >= '3.14' and platform_machine == 'arm64' and sys_platform == 'darwin'",
4785
+ "python_full_version < '3.14' and platform_machine == 'arm64' and sys_platform == 'darwin'",
4786
+ ]
4611
4787
  dependencies = [
4612
4788
  { name = "huggingface-hub", version = "0.36.2", source = { registry = "https://pypi.org/simple" }, marker = "extra == 'extra-14-extract-python-marker'" },
4613
- { name = "huggingface-hub", version = "1.33.0", source = { registry = "https://pypi.org/simple" }, marker = "extra == 'extra-14-extract-python-mineru' or extra != 'extra-14-extract-python-marker'" },
4789
+ { name = "huggingface-hub", version = "1.33.0", source = { registry = "https://pypi.org/simple" }, marker = "(sys_platform != 'darwin' and extra != 'extra-14-extract-python-marker') or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
4614
4790
  ]
4615
4791
  sdist = { url = "https://files.pythonhosted.org/packages/73/6f/f80cfef4a312e1fb34baf7d85c72d4411afde10978d4657f8cdd811d3ccc/tokenizers-0.22.2.tar.gz", hash = "sha256:473b83b915e547aa366d1eee11806deaf419e17be16310ac0a14077f1e28f917", size = 372115, upload-time = "2026-01-05T10:45:15.988Z" }
4616
4792
  wheels = [
@@ -4631,6 +4807,39 @@ wheels = [
4631
4807
  { url = "https://files.pythonhosted.org/packages/72/f4/0de46cfa12cdcbcd464cc59fde36912af405696f687e53a091fb432f694c/tokenizers-0.22.2-cp39-abi3-win_arm64.whl", hash = "sha256:9ce725d22864a1e965217204946f830c37876eee3b2ba6fc6255e8e903d5fcbc", size = 2612133, upload-time = "2026-01-05T10:45:17.232Z" },
4632
4808
  ]
4633
4809
 
4810
+ [[package]]
4811
+ name = "tokenizers"
4812
+ version = "0.23.2"
4813
+ source = { registry = "https://pypi.org/simple" }
4814
+ resolution-markers = [
4815
+ "python_full_version >= '3.14' and platform_machine != 'arm64' and sys_platform == 'darwin'",
4816
+ "python_full_version < '3.14' and platform_machine != 'arm64' and sys_platform == 'darwin'",
4817
+ "python_full_version >= '3.14' and platform_machine == 'arm64' and sys_platform == 'darwin'",
4818
+ "python_full_version < '3.14' and platform_machine == 'arm64' and sys_platform == 'darwin'",
4819
+ ]
4820
+ dependencies = [
4821
+ { name = "huggingface-hub", version = "1.33.0", source = { registry = "https://pypi.org/simple" }, marker = "(sys_platform == 'darwin' and extra != 'extra-14-extract-python-marker') or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
4822
+ ]
4823
+ sdist = { url = "https://files.pythonhosted.org/packages/18/1e/bc6587c5ab643b2e17776cace9070a2ae73549c86bffac9934a600bf3c31/tokenizers-0.23.2.tar.gz", hash = "sha256:7f0f085686b9de0d0079e6f874ae053600db64c5d13049e0bbc0119926d25aac", size = 385745, upload-time = "2026-09-03T08:55:42.89Z" }
4824
+ wheels = [
4825
+ { url = "https://files.pythonhosted.org/packages/4d/ed/8a443528baa6fac8dfe8c3b75b038c63ac92bb539bcabe311e227c718173/tokenizers-0.23.2-cp310-abi3-macosx_10_12_x86_64.whl", hash = "sha256:85a9a357a3764aecc904ee76bdaf8cf1ad8e5a67a1b929a487c4a39b49ed0e90", size = 3148852, upload-time = "2026-09-03T08:55:30.874Z" },
4826
+ { url = "https://files.pythonhosted.org/packages/67/49/22da045a91732384d3a3771816bf188dc5a1f702c32e635afa7c679c0bef/tokenizers-0.23.2-cp310-abi3-macosx_11_0_arm64.whl", hash = "sha256:986670e43691469dcee610ea0f846f91a8f84e91fc6f7a48d4c064414c0ec2bf", size = 3101593, upload-time = "2026-09-03T08:55:28.587Z" },
4827
+ { url = "https://files.pythonhosted.org/packages/2e/4d/8f569ed49372a3ed8e57099bd515055fd48d7c95912c4307cda6973c2168/tokenizers-0.23.2-cp310-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a37039b5dfc4af84eb3ef0a92f4307e28936c8f9adccba2629d36f652e9bf7a2", size = 3516830, upload-time = "2026-09-03T08:55:14.741Z" },
4828
+ { url = "https://files.pythonhosted.org/packages/2a/de/e2f14c8919d5bf51874051d00d6c7b7e0e8bde6c6a2dbeddda7f642896ff/tokenizers-0.23.2-cp310-abi3-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:7b7e37ba198f24150f523e1242e83c4970de4a525480586be5dcc24d9add32c5", size = 3407975, upload-time = "2026-09-03T08:55:16.842Z" },
4829
+ { url = "https://files.pythonhosted.org/packages/c5/bd/93c69152d02ef06ce47aed8b2bf4952dcf733c935a62791873932b2934d9/tokenizers-0.23.2-cp310-abi3-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:43e4f2071e3cc8d5d86421c874aebc82659bb51a68bcdef5a0da75ee89511ccb", size = 3748165, upload-time = "2026-09-03T08:55:24.769Z" },
4830
+ { url = "https://files.pythonhosted.org/packages/2d/b7/56b84b80bc96942bba8eb23751a9e8a1fce4faaf4390425e7083f721c98c/tokenizers-0.23.2-cp310-abi3-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:325fee2e0418a9dc6c9ecf736a5f5f0db7875183ace9549ae339da76f7a1fbb7", size = 4024165, upload-time = "2026-09-03T08:55:18.806Z" },
4831
+ { url = "https://files.pythonhosted.org/packages/9b/8a/0175e216f005c2fe08238292663aa41e4c802b216e71047a69a0e9fc6fa3/tokenizers-0.23.2-cp310-abi3-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:950d7c9426fa72406a0ffeacdbc0bb9985f5db20eb8b263f29c79aaf83105703", size = 3591899, upload-time = "2026-09-03T08:55:22.752Z" },
4832
+ { url = "https://files.pythonhosted.org/packages/2c/ca/ca6b93c7820df123b2662a9469e8facc826ccc94e98fdd0d615f6431e73a/tokenizers-0.23.2-cp310-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:41c2f84d172449b4dadb9cdc508e3e364076613c35b16e76ecfe47a60d1e3305", size = 3386843, upload-time = "2026-09-03T08:55:26.584Z" },
4833
+ { url = "https://files.pythonhosted.org/packages/e9/a4/4f9106d317b14a80aefea9f0e3a8d07ef25f856a7607eb7f5ab894281fcb/tokenizers-0.23.2-cp310-abi3-manylinux_2_31_riscv64.whl", hash = "sha256:12f0835dc2ee694746a76adf7b1567d4346a4a502ebe93fb1f5f80ea49799b78", size = 3577314, upload-time = "2026-09-03T08:55:20.825Z" },
4834
+ { url = "https://files.pythonhosted.org/packages/8d/6a/1552b70fb0d9ab074fd3fc961435d01364e79c9058481822c3af6e8d402c/tokenizers-0.23.2-cp310-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:eb2f9c8a24da020ea8c11a01a19c1c2547912d92121ae4a01cfbca46125dee40", size = 9967367, upload-time = "2026-09-03T08:55:33.188Z" },
4835
+ { url = "https://files.pythonhosted.org/packages/06/01/3ccb3a956c7528b2507b8a9714155c4baf86af593039db6ea375dd0c96c3/tokenizers-0.23.2-cp310-abi3-musllinux_1_2_armv7l.whl", hash = "sha256:f486f402f6f9abee5bb032553736813af0c710a86b2e0ca592634c55cea1f835", size = 9811886, upload-time = "2026-09-03T08:55:35.642Z" },
4836
+ { url = "https://files.pythonhosted.org/packages/fa/73/7038e612d48bda1599457f712f6bd3854eae1a9dc9c13aa47f835349db48/tokenizers-0.23.2-cp310-abi3-musllinux_1_2_i686.whl", hash = "sha256:bef235815a067b2648caf6dcc7a71091b0b0fff9ee8057f6451eb9335fae52ef", size = 10146224, upload-time = "2026-09-03T08:55:38.391Z" },
4837
+ { url = "https://files.pythonhosted.org/packages/b5/d8/8e9e4e0b287a338d8f88976729628c9d22e8a54cfaf9777018a7f7cb58a0/tokenizers-0.23.2-cp310-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:5c56bda1511921587789163e524d196ed8284174ac23abd7685d5ea8da6c4718", size = 10256304, upload-time = "2026-09-03T08:55:40.977Z" },
4838
+ { url = "https://files.pythonhosted.org/packages/f3/1f/c79a01f671a49728ebb0b61f7ff9ea45663b66cab40bc0858e9859b25c16/tokenizers-0.23.2-cp310-abi3-win32.whl", hash = "sha256:debf978920d93ba9c219bd67cc4bbfaf912c9039e41e7a28b91ec15e3728c95a", size = 2592809, upload-time = "2026-09-03T08:55:48.02Z" },
4839
+ { url = "https://files.pythonhosted.org/packages/db/f7/0a69ac6b82dbccf3f71add938a161c497952749294b8dd6dfe03a819dc40/tokenizers-0.23.2-cp310-abi3-win_amd64.whl", hash = "sha256:2e96f5699d5249c9c64aa8412e044f727aae3a4098cf830f9901ec1afc361cde", size = 2863236, upload-time = "2026-09-03T08:55:46.193Z" },
4840
+ { url = "https://files.pythonhosted.org/packages/d7/b0/dee84cb44175be1b4c35bd2f770727494e78f0bb38e571a623ade94dbebb/tokenizers-0.23.2-cp310-abi3-win_arm64.whl", hash = "sha256:e49c394456dd9985787fec76132438ba3fb8911f857b1bf3d40119f9292d41aa", size = 2729352, upload-time = "2026-09-03T08:55:44.345Z" },
4841
+ ]
4842
+
4634
4843
  [[package]]
4635
4844
  name = "tomlkit"
4636
4845
  version = "0.14.0"
@@ -4830,7 +5039,7 @@ dependencies = [
4830
5039
  { name = "regex", version = "2024.11.6", source = { registry = "https://pypi.org/simple" }, marker = "extra == 'extra-14-extract-python-marker'" },
4831
5040
  { name = "requests", marker = "extra == 'extra-14-extract-python-marker'" },
4832
5041
  { name = "safetensors", marker = "extra == 'extra-14-extract-python-marker'" },
4833
- { name = "tokenizers", marker = "extra == 'extra-14-extract-python-marker'" },
5042
+ { name = "tokenizers", version = "0.22.2", source = { registry = "https://pypi.org/simple" }, marker = "extra == 'extra-14-extract-python-marker'" },
4834
5043
  { name = "tqdm", marker = "extra == 'extra-14-extract-python-marker'" },
4835
5044
  ]
4836
5045
  sdist = { url = "https://files.pythonhosted.org/packages/c4/35/67252acc1b929dc88b6602e8c4a982e64f31e733b804c14bc24b47da35e6/transformers-4.57.6.tar.gz", hash = "sha256:55e44126ece9dc0a291521b7e5492b572e6ef2766338a610b9ab5afbb70689d3", size = 10134912, upload-time = "2026-01-16T10:38:39.284Z" }
@@ -4845,33 +5054,55 @@ source = { registry = "https://pypi.org/simple" }
4845
5054
  resolution-markers = [
4846
5055
  "python_full_version >= '3.14' and sys_platform == 'win32'",
4847
5056
  "python_full_version >= '3.14' and sys_platform == 'emscripten'",
4848
- "python_full_version >= '3.14' and platform_machine != 'arm64' and sys_platform == 'darwin'",
4849
5057
  "python_full_version >= '3.14' and platform_machine == 'aarch64' and sys_platform == 'linux'",
4850
5058
  "(python_full_version >= '3.14' and platform_machine != 'aarch64' and sys_platform == 'linux') or (python_full_version >= '3.14' and sys_platform != 'darwin' and sys_platform != 'emscripten' and sys_platform != 'linux' and sys_platform != 'win32')",
4851
5059
  "python_full_version < '3.14' and sys_platform == 'win32'",
4852
5060
  "python_full_version < '3.14' and sys_platform == 'emscripten'",
4853
- "python_full_version < '3.14' and platform_machine != 'arm64' and sys_platform == 'darwin'",
4854
5061
  "python_full_version < '3.14' and platform_machine == 'aarch64' and sys_platform == 'linux'",
4855
5062
  "(python_full_version < '3.14' and platform_machine != 'aarch64' and sys_platform == 'linux') or (python_full_version < '3.14' and sys_platform != 'darwin' and sys_platform != 'emscripten' and sys_platform != 'linux' and sys_platform != 'win32')",
4856
- "python_full_version >= '3.14' and platform_machine == 'arm64' and sys_platform == 'darwin'",
4857
- "python_full_version < '3.14' and platform_machine == 'arm64' and sys_platform == 'darwin'",
4858
5063
  ]
4859
5064
  dependencies = [
4860
- { name = "huggingface-hub", version = "1.33.0", source = { registry = "https://pypi.org/simple" }, marker = "extra == 'extra-14-extract-python-mineru' or extra != 'extra-14-extract-python-marker'" },
4861
- { name = "numpy", marker = "extra == 'extra-14-extract-python-mineru' or extra != 'extra-14-extract-python-marker'" },
4862
- { name = "packaging", marker = "extra == 'extra-14-extract-python-mineru' or extra != 'extra-14-extract-python-marker'" },
4863
- { name = "pyyaml", marker = "extra == 'extra-14-extract-python-mineru' or extra != 'extra-14-extract-python-marker'" },
4864
- { name = "regex", version = "2026.9.29", source = { registry = "https://pypi.org/simple" }, marker = "extra == 'extra-14-extract-python-mineru' or extra != 'extra-14-extract-python-marker'" },
4865
- { name = "safetensors", marker = "extra == 'extra-14-extract-python-mineru' or extra != 'extra-14-extract-python-marker'" },
4866
- { name = "tokenizers", marker = "extra == 'extra-14-extract-python-mineru' or extra != 'extra-14-extract-python-marker'" },
4867
- { name = "tqdm", marker = "extra == 'extra-14-extract-python-mineru' or extra != 'extra-14-extract-python-marker'" },
4868
- { name = "typer", marker = "extra == 'extra-14-extract-python-mineru' or extra != 'extra-14-extract-python-marker'" },
5065
+ { name = "huggingface-hub", version = "1.33.0", source = { registry = "https://pypi.org/simple" }, marker = "(sys_platform != 'darwin' and extra != 'extra-14-extract-python-marker') or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
5066
+ { name = "numpy", marker = "(sys_platform != 'darwin' and extra != 'extra-14-extract-python-marker') or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
5067
+ { name = "packaging", marker = "(sys_platform != 'darwin' and extra != 'extra-14-extract-python-marker') or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
5068
+ { name = "pyyaml", marker = "(sys_platform != 'darwin' and extra != 'extra-14-extract-python-marker') or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
5069
+ { name = "regex", version = "2026.9.29", source = { registry = "https://pypi.org/simple" }, marker = "(sys_platform != 'darwin' and extra != 'extra-14-extract-python-marker') or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
5070
+ { name = "safetensors", marker = "(sys_platform != 'darwin' and extra != 'extra-14-extract-python-marker') or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
5071
+ { name = "tokenizers", version = "0.22.2", source = { registry = "https://pypi.org/simple" }, marker = "(sys_platform != 'darwin' and extra != 'extra-14-extract-python-marker') or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
5072
+ { name = "tqdm", marker = "(sys_platform != 'darwin' and extra != 'extra-14-extract-python-marker') or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
5073
+ { name = "typer", marker = "(sys_platform != 'darwin' and extra != 'extra-14-extract-python-marker') or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
4869
5074
  ]
4870
5075
  sdist = { url = "https://files.pythonhosted.org/packages/2a/92/c50c61da7046bbb59a4d011291aeadcfb4d7980ab36fdb31e93823a3fb93/transformers-5.15.1.tar.gz", hash = "sha256:27c996bd9075ddc82d40f8590dfdc81ea45f611bfca477e0db5d7fd257a482f7", size = 9378434, upload-time = "2026-08-19T11:28:20.33Z" }
4871
5076
  wheels = [
4872
5077
  { url = "https://files.pythonhosted.org/packages/41/c4/a12e1d9b387fb0c40a57116db82b457e8c771cb419163cda29204d74a595/transformers-5.15.1-py3-none-any.whl", hash = "sha256:b7cdf238ff583e3a58dbc7fa34da1aaf091ce063141f65a30538160bd5afe93f", size = 11749582, upload-time = "2026-08-19T11:28:16.726Z" },
4873
5078
  ]
4874
5079
 
5080
+ [[package]]
5081
+ name = "transformers"
5082
+ version = "5.18.0"
5083
+ source = { registry = "https://pypi.org/simple" }
5084
+ resolution-markers = [
5085
+ "python_full_version >= '3.14' and platform_machine != 'arm64' and sys_platform == 'darwin'",
5086
+ "python_full_version < '3.14' and platform_machine != 'arm64' and sys_platform == 'darwin'",
5087
+ "python_full_version >= '3.14' and platform_machine == 'arm64' and sys_platform == 'darwin'",
5088
+ "python_full_version < '3.14' and platform_machine == 'arm64' and sys_platform == 'darwin'",
5089
+ ]
5090
+ dependencies = [
5091
+ { name = "huggingface-hub", version = "1.33.0", source = { registry = "https://pypi.org/simple" }, marker = "(sys_platform == 'darwin' and extra != 'extra-14-extract-python-marker') or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
5092
+ { name = "numpy", marker = "(sys_platform == 'darwin' and extra != 'extra-14-extract-python-marker') or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
5093
+ { name = "packaging", marker = "(sys_platform == 'darwin' and extra != 'extra-14-extract-python-marker') or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
5094
+ { name = "pyyaml", marker = "(sys_platform == 'darwin' and extra != 'extra-14-extract-python-marker') or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
5095
+ { name = "regex", version = "2026.9.29", source = { registry = "https://pypi.org/simple" }, marker = "(sys_platform == 'darwin' and extra != 'extra-14-extract-python-marker') or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
5096
+ { name = "safetensors", marker = "(sys_platform == 'darwin' and extra != 'extra-14-extract-python-marker') or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
5097
+ { name = "tokenizers", version = "0.23.2", source = { registry = "https://pypi.org/simple" }, marker = "(sys_platform == 'darwin' and extra != 'extra-14-extract-python-marker') or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
5098
+ { name = "tqdm", marker = "(sys_platform == 'darwin' and extra != 'extra-14-extract-python-marker') or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
5099
+ { name = "typer", marker = "(sys_platform == 'darwin' and extra != 'extra-14-extract-python-marker') or (extra == 'extra-14-extract-python-marker' and extra == 'extra-14-extract-python-mineru')" },
5100
+ ]
5101
+ sdist = { url = "https://files.pythonhosted.org/packages/9e/6e/5a50ca8aff5fdd4ff21271198d2fb9e1f11bcd0ea279576199c0caa34cba/transformers-5.18.0.tar.gz", hash = "sha256:d89c206e42e841af7cd314b3d2b21c03e977d819cdb4aef65ada048769a644df", size = 10028745, upload-time = "2026-09-30T16:46:00.575Z" }
5102
+ wheels = [
5103
+ { url = "https://files.pythonhosted.org/packages/2d/3b/f84d7d12886147c5267216ed6c57d1163e68d24a89b2ad29d835a7089463/transformers-5.18.0-py3-none-any.whl", hash = "sha256:d79e5a515a572ee3deb33eb0d578be47d91e70c75d54cf21bf6eede6edef14a9", size = 12579646, upload-time = "2026-09-30T16:45:57.51Z" },
5104
+ ]
5105
+
4875
5106
  [[package]]
4876
5107
  name = "tree-sitter"
4877
5108
  version = "0.25.2"
@@ -4959,7 +5190,7 @@ wheels = [
4959
5190
 
4960
5191
  [[package]]
4961
5192
  name = "typer"
4962
- version = "0.24.2"
5193
+ version = "0.21.2"
4963
5194
  source = { registry = "https://pypi.org/simple" }
4964
5195
  dependencies = [
4965
5196
  { name = "annotated-doc" },
@@ -4967,9 +5198,9 @@ dependencies = [
4967
5198
  { name = "rich" },
4968
5199
  { name = "shellingham" },
4969
5200
  ]
4970
- sdist = { url = "https://files.pythonhosted.org/packages/83/b8/9ebb531b6c2d377af08ac6746a5df3425b21853a5d2260876919b58a2a4a/typer-0.24.2.tar.gz", hash = "sha256:ec070dcfca1408e85ee203c6365001e818c3b7fffe686fd07ff2d68095ca0480", size = 119849, upload-time = "2026-04-22T17:45:34.413Z" }
5201
+ sdist = { url = "https://files.pythonhosted.org/packages/f2/1e/a27cc02a0cd715118c71fa2aef2c687fdefc3c28d90fd0dd789c5118154c/typer-0.21.2.tar.gz", hash = "sha256:1abd95a3b675e17ff61b0838ac637fe9478d446d62ad17fa4bb81ea57cc54028", size = 120426, upload-time = "2026-02-10T19:33:46.182Z" }
4971
5202
  wheels = [
4972
- { url = "https://files.pythonhosted.org/packages/39/d1/9484b497e0a0410b901c12b8251c3e746e1e863f7d28419ffe06f7892fda/typer-0.24.2-py3-none-any.whl", hash = "sha256:b618bc3d721f9a8d30f3e05565be26416d06e9bcc29d49bc491dc26aba674fa8", size = 55977, upload-time = "2026-04-22T17:45:33.055Z" },
5203
+ { url = "https://files.pythonhosted.org/packages/b8/cc/d59f893fbdfb5f58770c05febfc4086a46875f1084453621c35605cec946/typer-0.21.2-py3-none-any.whl", hash = "sha256:c3d8de54d00347ef90b82131ca946274f017cffb46683ae3883c360fa958f55c", size = 56728, upload-time = "2026-02-10T19:33:48.01Z" },
4973
5204
  ]
4974
5205
 
4975
5206
  [[package]]