docling 2.51.0__tar.gz → 2.52.0__tar.gz

{docling-2.51.0 → docling-2.52.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: docling
-Version: 2.51.0
+Version: 2.52.0
 Summary: SDK and CLI for parsing PDF, DOCX, HTML, and more, to a unified document representation for powering downstream workflows such as gen AI applications.
 Author-email: Christoph Auer <cau@zurich.ibm.com>, Michele Dolfi <dol@zurich.ibm.com>, Maxim Lysak <mly@zurich.ibm.com>, Nikos Livathinos <nli@zurich.ibm.com>, Ahmed Nassar <ahn@zurich.ibm.com>, Panos Vagenas <pva@zurich.ibm.com>, Peter Staar <taa@zurich.ibm.com>
 License-Expression: MIT
@@ -26,7 +26,7 @@ Requires-Python: <4.0,>=3.9
 Description-Content-Type: text/markdown
 License-File: LICENSE
 Requires-Dist: pydantic<3.0.0,>=2.0.0
-Requires-Dist: docling-core[chunking]<3.0.0,>=2.42.0
+Requires-Dist: docling-core[chunking]<3.0.0,>=2.48.0
 Requires-Dist: docling-parse<5.0.0,>=4.4.0
 Requires-Dist: docling-ibm-models<4,>=3.9.1
 Requires-Dist: filetype<2.0.0,>=1.2.0
@@ -110,16 +110,20 @@ Docling simplifies document processing, parsing diverse formats — including ad
 * 🔍 Extensive OCR support for scanned PDFs and images
 * 👓 Support of several Visual Language Models ([SmolDocling](https://huggingface.co/ds4sd/SmolDocling-256M-preview))
 * 🎙️ Audio support with Automatic Speech Recognition (ASR) models
+* 🔌 Connect to any agent using the [MCP server](https://docling-project.github.io/docling/usage/mcp/)
 * 💻 Simple and convenient CLI
 
 ### What's new
 * 📤 Structured [information extraction][extraction] \[🧪 beta\]
+* 📑 New layout model (**Heron**) by default, for faster PDF parsing
+* 🔌 [MCP server](https://docling-project.github.io/docling/usage/mcp/) for agentic applications
 
 ### Coming soon
 
 * 📝 Metadata extraction, including title, authors, references & language
 * 📝 Chart understanding (Barchart, Piechart, LinePlot, etc)
 * 📝 Complex chemistry understanding (Molecular structures)
+* 📝 Parsing of Web Video Text Tracks (WebVTT) files
 
 ## Installation
 
@@ -145,7 +149,7 @@ result = converter.convert(source)
 print(result.document.export_to_markdown())  # output: "## Docling Technical Report[...]"
 ```
 
-More [advanced usage options](https://docling-project.github.io/docling/usage/) are available in
+More [advanced usage options](https://docling-project.github.io/docling/usage/advanced_options/) are available in
 the docs.
 
 ## CLI

{docling-2.51.0 → docling-2.52.0}/README.md

@@ -38,16 +38,20 @@ Docling simplifies document processing, parsing diverse formats — including ad
 * 🔍 Extensive OCR support for scanned PDFs and images
 * 👓 Support of several Visual Language Models ([SmolDocling](https://huggingface.co/ds4sd/SmolDocling-256M-preview))
 * 🎙️ Audio support with Automatic Speech Recognition (ASR) models
+* 🔌 Connect to any agent using the [MCP server](https://docling-project.github.io/docling/usage/mcp/)
 * 💻 Simple and convenient CLI
 
 ### What's new
 * 📤 Structured [information extraction][extraction] \[🧪 beta\]
+* 📑 New layout model (**Heron**) by default, for faster PDF parsing
+* 🔌 [MCP server](https://docling-project.github.io/docling/usage/mcp/) for agentic applications
 
 ### Coming soon
 
 * 📝 Metadata extraction, including title, authors, references & language
 * 📝 Chart understanding (Barchart, Piechart, LinePlot, etc)
 * 📝 Complex chemistry understanding (Molecular structures)
+* 📝 Parsing of Web Video Text Tracks (WebVTT) files
 
 ## Installation
 
@@ -73,7 +77,7 @@ result = converter.convert(source)
 print(result.document.export_to_markdown())  # output: "## Docling Technical Report[...]"
 ```
 
-More [advanced usage options](https://docling-project.github.io/docling/usage/) are available in
+More [advanced usage options](https://docling-project.github.io/docling/usage/advanced_options/) are available in
 the docs.
 
 ## CLI

{docling-2.51.0 → docling-2.52.0}/docling/cli/main.py

@@ -48,6 +48,7 @@ from docling.datamodel.base_models import (
 from docling.datamodel.document import ConversionResult
 from docling.datamodel.pipeline_options import (
     AsrPipelineOptions,
+    ConvertPipelineOptions,
     EasyOcrOptions,
     OcrOptions,
     PaginatedPipelineOptions,
@@ -71,8 +72,13 @@ from docling.datamodel.vlm_model_specs import (
 from docling.document_converter import (
     AudioFormatOption,
     DocumentConverter,
+    ExcelFormatOption,
     FormatOption,
+    HTMLFormatOption,
+    MarkdownFormatOption,
     PdfFormatOption,
+    PowerpointFormatOption,
+    WordFormatOption,
 )
 from docling.models.factories import get_ocr_factory
 from docling.pipeline.asr_pipeline import AsrPipeline
@@ -626,10 +632,33 @@ def convert(  # noqa: C901
                 backend=MetsGbsDocumentBackend,
             )
 
+            # SimplePipeline options
+            simple_format_option = ConvertPipelineOptions(
+                do_picture_description=enrich_picture_description,
+                do_picture_classification=enrich_picture_classes,
+            )
+            if artifacts_path is not None:
+                simple_format_option.artifacts_path = artifacts_path
+
             format_options = {
                 InputFormat.PDF: pdf_format_option,
                 InputFormat.IMAGE: pdf_format_option,
                 InputFormat.METS_GBS: mets_gbs_format_option,
+                InputFormat.DOCX: WordFormatOption(
+                    pipeline_options=simple_format_option
+                ),
+                InputFormat.PPTX: PowerpointFormatOption(
+                    pipeline_options=simple_format_option
+                ),
+                InputFormat.XLSX: ExcelFormatOption(
+                    pipeline_options=simple_format_option
+                ),
+                InputFormat.HTML: HTMLFormatOption(
+                    pipeline_options=simple_format_option
+                ),
+                InputFormat.MD: MarkdownFormatOption(
+                    pipeline_options=simple_format_option
+                ),
             }
 
         elif pipeline == ProcessingPipeline.VLM:

{docling-2.51.0 → docling-2.52.0}/docling/datamodel/pipeline_options.py

@@ -135,6 +135,8 @@ class EasyOcrOptions(OcrOptions):
     recog_network: Optional[str] = "standard"
     download_enabled: bool = True
 
+    suppress_mps_warnings: bool = True
+
     model_config = ConfigDict(
         extra="forbid",
         protected_namespaces=(),
@@ -257,11 +259,21 @@ class PipelineOptions(BaseOptions):
     accelerator_options: AcceleratorOptions = AcceleratorOptions()
     enable_remote_services: bool = False
     allow_external_plugins: bool = False
+    artifacts_path: Optional[Union[Path, str]] = None
 
 
-class PaginatedPipelineOptions(PipelineOptions):
-    artifacts_path: Optional[Union[Path, str]] = None
+class ConvertPipelineOptions(PipelineOptions):
+    """Base convert pipeline options."""
+
+    do_picture_classification: bool = False  # True: classify pictures in documents
+
+    do_picture_description: bool = False  # True: run describe pictures in documents
+    picture_description_options: PictureDescriptionBaseOptions = (
+        smolvlm_picture_description
+    )
+
 
+class PaginatedPipelineOptions(ConvertPipelineOptions):
     images_scale: float = 1.0
     generate_page_images: bool = False
     generate_picture_images: bool = False
@@ -293,13 +305,11 @@ class LayoutOptions(BaseModel):
 
 class AsrPipelineOptions(PipelineOptions):
     asr_options: Union[InlineAsrOptions] = asr_model_specs.WHISPER_TINY
-    artifacts_path: Optional[Union[Path, str]] = None
 
 
 class VlmExtractionPipelineOptions(PipelineOptions):
     """Options for extraction pipeline."""
 
-    artifacts_path: Optional[Union[Path, str]] = None
     vlm_options: Union[InlineVlmOptions] = NU_EXTRACT_2B_TRANSFORMERS
 
 
@@ -310,8 +320,6 @@ class PdfPipelineOptions(PaginatedPipelineOptions):
     do_ocr: bool = True  # True: perform OCR, replace programmatic PDF text
     do_code_enrichment: bool = False  # True: perform code OCR
     do_formula_enrichment: bool = False  # True: perform formula OCR, return Latex code
-    do_picture_classification: bool = False  # True: classify pictures in documents
-    do_picture_description: bool = False  # True: run describe pictures in documents
     force_backend_text: bool = (
         False  # (To be used with vlms, or other generative models)
     )
@@ -319,9 +327,6 @@ class PdfPipelineOptions(PaginatedPipelineOptions):
 
     table_structure_options: TableStructureOptions = TableStructureOptions()
     ocr_options: OcrOptions = EasyOcrOptions()
-    picture_description_options: PictureDescriptionBaseOptions = (
-        smolvlm_picture_description
-    )
     layout_options: LayoutOptions = LayoutOptions()
 
     images_scale: float = 1.0

{docling-2.51.0 → docling-2.52.0}/docling/models/base_model.py

@@ -4,7 +4,13 @@ from collections.abc import Iterable
 from typing import Any, Generic, Optional, Protocol, Type, Union
 
 import numpy as np
-from docling_core.types.doc import BoundingBox, DocItem, DoclingDocument, NodeItem
+from docling_core.types.doc import (
+    BoundingBox,
+    DocItem,
+    DoclingDocument,
+    NodeItem,
+    PictureItem,
+)
 from PIL.Image import Image
 from typing_extensions import TypeVar
 
@@ -164,8 +170,17 @@ class BaseItemAndImageEnrichmentModel(
             return None
 
         assert isinstance(element, DocItem)
-        element_prov = element.prov[0]
 
+        # Allow the case of documents without page images but embedded images (e.g. Word and HTML docs)
+        if len(element.prov) == 0 and isinstance(element, PictureItem):
+            embedded_im = element.get_image(conv_res.document)
+            if embedded_im is not None:
+                return ItemAndImageEnrichmentElement(item=element, image=embedded_im)
+            else:
+                return None
+
+        # Crop the image form the page
+        element_prov = element.prov[0]
         bbox = element_prov.bbox
         width = bbox.r - bbox.l
         height = bbox.t - bbox.b
@@ -183,4 +198,14 @@ class BaseItemAndImageEnrichmentModel(
         cropped_image = conv_res.pages[page_ix].get_image(
             scale=self.images_scale, cropbox=expanded_bbox
         )
+
+        # Allow for images being embedded without the page backend or page images
+        if cropped_image is None and isinstance(element, PictureItem):
+            embedded_im = element.get_image(conv_res.document)
+            if embedded_im is not None:
+                return ItemAndImageEnrichmentElement(item=element, image=embedded_im)
+            else:
+                return None
+
+        # Return the proper cropped image
         return ItemAndImageEnrichmentElement(item=element, image=cropped_image)

{docling-2.51.0 → docling-2.52.0}/docling/models/easyocr_model.py

@@ -78,14 +78,17 @@ class EasyOcrModel(BaseOcrModel):
                 download_enabled = False
                 model_storage_directory = str(artifacts_path / self._model_repo_folder)
 
-            self.reader = easyocr.Reader(
-                lang_list=self.options.lang,
-                gpu=use_gpu,
-                model_storage_directory=model_storage_directory,
-                recog_network=self.options.recog_network,
-                download_enabled=download_enabled,
-                verbose=False,
-            )
+            with warnings.catch_warnings():
+                if self.options.suppress_mps_warnings:
+                    warnings.filterwarnings("ignore", message=".*pin_memory.*MPS.*")
+                self.reader = easyocr.Reader(
+                    lang_list=self.options.lang,
+                    gpu=use_gpu,
+                    model_storage_directory=model_storage_directory,
+                    recog_network=self.options.recog_network,
+                    download_enabled=download_enabled,
+                    verbose=False,
+                )
 
     @staticmethod
     def download_models(
@@ -147,7 +150,14 @@ class EasyOcrModel(BaseOcrModel):
                             scale=self.scale, cropbox=ocr_rect
                         )
                         im = numpy.array(high_res_image)
-                        result = self.reader.readtext(im)
+
+                        with warnings.catch_warnings():
+                            if self.options.suppress_mps_warnings:
+                                warnings.filterwarnings(
+                                    "ignore", message=".*pin_memory.*MPS.*"
+                                )
+
+                            result = self.reader.readtext(im)
 
                         del high_res_image
                         del im

{docling-2.51.0 → docling-2.52.0}/docling/models/picture_description_vlm_model.py

@@ -67,7 +67,7 @@ class PictureDescriptionVlmModel(
                 self.model = AutoModelForImageTextToText.from_pretrained(
                     artifacts_path,
                     device_map=self.device,
-                    torch_dtype=torch.bfloat16,
+                    dtype=torch.bfloat16,
                     _attn_implementation=(
                         "flash_attention_2"
                         if self.device.startswith("cuda")

{docling-2.51.0 → docling-2.52.0}/docling/models/vlm_models_inline/hf_transformers_model.py

@@ -112,7 +112,7 @@ class HuggingFaceTransformersVlmModel(BaseVlmPageModel, HuggingFaceModelDownload
             self.vlm_model = model_cls.from_pretrained(
                 artifacts_path,
                 device_map=self.device,
-                torch_dtype=self.vlm_options.torch_dtype,
+                dtype=self.vlm_options.torch_dtype,
                 _attn_implementation=(
                     "flash_attention_2"
                     if self.device.startswith("cuda")

{docling-2.51.0 → docling-2.52.0}/docling/models/vlm_models_inline/nuextract_transformers_model.py

@@ -144,7 +144,7 @@ class NuExtractTransformersModel(BaseVlmModel, HuggingFaceModelDownloadMixin):
             self.vlm_model = AutoModelForImageTextToText.from_pretrained(
                 artifacts_path,
                 device_map=self.device,
-                torch_dtype=self.vlm_options.torch_dtype,
+                dtype=self.vlm_options.torch_dtype,
                 _attn_implementation=(
                     "flash_attention_2"
                     if self.device.startswith("cuda")

{docling-2.51.0 → docling-2.52.0}/docling/pipeline/asr_pipeline.py

@@ -208,25 +208,13 @@ class AsrPipeline(BasePipeline):
 
         self.pipeline_options: AsrPipelineOptions = pipeline_options
 
-        artifacts_path: Optional[Path] = None
-        if pipeline_options.artifacts_path is not None:
-            artifacts_path = Path(pipeline_options.artifacts_path).expanduser()
-        elif settings.artifacts_path is not None:
-            artifacts_path = Path(settings.artifacts_path).expanduser()
-
-        if artifacts_path is not None and not artifacts_path.is_dir():
-            raise RuntimeError(
-                f"The value of {artifacts_path=} is not valid. "
-                "When defined, it must point to a folder containing all models required by the pipeline."
-            )
-
         if isinstance(self.pipeline_options.asr_options, InlineAsrNativeWhisperOptions):
             asr_options: InlineAsrNativeWhisperOptions = (
                 self.pipeline_options.asr_options
             )
             self._model = _NativeWhisperModel(
                 enabled=True,  # must be always enabled for this pipeline to make sense.
-                artifacts_path=artifacts_path,
+                artifacts_path=self.artifacts_path,
                 accelerator_options=pipeline_options.accelerator_options,
                 asr_options=asr_options,
             )

{docling-2.51.0 → docling-2.52.0}/docling/pipeline/base_extraction_pipeline.py

@@ -1,19 +1,33 @@
 import logging
 from abc import ABC, abstractmethod
+from pathlib import Path
 from typing import Optional
 
 from docling.datamodel.base_models import ConversionStatus, ErrorItem
 from docling.datamodel.document import InputDocument
 from docling.datamodel.extraction import ExtractionResult, ExtractionTemplateType
-from docling.datamodel.pipeline_options import BaseOptions
+from docling.datamodel.pipeline_options import BaseOptions, PipelineOptions
+from docling.datamodel.settings import settings
 
 _log = logging.getLogger(__name__)
 
 
 class BaseExtractionPipeline(ABC):
-    def __init__(self, pipeline_options: BaseOptions):
+    def __init__(self, pipeline_options: PipelineOptions):
         self.pipeline_options = pipeline_options
 
+        self.artifacts_path: Optional[Path] = None
+        if pipeline_options.artifacts_path is not None:
+            self.artifacts_path = Path(pipeline_options.artifacts_path).expanduser()
+        elif settings.artifacts_path is not None:
+            self.artifacts_path = Path(settings.artifacts_path).expanduser()
+
+        if self.artifacts_path is not None and not self.artifacts_path.is_dir():
+            raise RuntimeError(
+                f"The value of {self.artifacts_path=} is not valid. "
+                "When defined, it must point to a folder containing all models required by the pipeline."
+            )
+
     def execute(
         self,
         in_doc: InputDocument,
@@ -54,5 +68,5 @@ class BaseExtractionPipeline(ABC):
 
     @classmethod
     @abstractmethod
-    def get_default_options(cls) -> BaseOptions:
+    def get_default_options(cls) -> PipelineOptions:
         pass

{docling-2.51.0 → docling-2.52.0}/docling/pipeline/base_pipeline.py

@@ -4,7 +4,8 @@ import time
 import traceback
 from abc import ABC, abstractmethod
 from collections.abc import Iterable
-from typing import Any, Callable, List
+from pathlib import Path
+from typing import Any, Callable, List, Optional
 
 from docling_core.types.doc import NodeItem
 
@@ -20,9 +21,19 @@ from docling.datamodel.base_models import (
     Page,
 )
 from docling.datamodel.document import ConversionResult, InputDocument
-from docling.datamodel.pipeline_options import PdfPipelineOptions, PipelineOptions
+from docling.datamodel.pipeline_options import (
+    ConvertPipelineOptions,
+    PdfPipelineOptions,
+    PipelineOptions,
+)
 from docling.datamodel.settings import settings
 from docling.models.base_model import GenericEnrichmentModel
+from docling.models.document_picture_classifier import (
+    DocumentPictureClassifier,
+    DocumentPictureClassifierOptions,
+)
+from docling.models.factories import get_picture_description_factory
+from docling.models.picture_description_base_model import PictureDescriptionBaseModel
 from docling.utils.profiling import ProfilingScope, TimeRecorder
 from docling.utils.utils import chunkify
 
@@ -36,6 +47,18 @@ class BasePipeline(ABC):
         self.build_pipe: List[Callable] = []
         self.enrichment_pipe: List[GenericEnrichmentModel[Any]] = []
 
+        self.artifacts_path: Optional[Path] = None
+        if pipeline_options.artifacts_path is not None:
+            self.artifacts_path = Path(pipeline_options.artifacts_path).expanduser()
+        elif settings.artifacts_path is not None:
+            self.artifacts_path = Path(settings.artifacts_path).expanduser()
+
+        if self.artifacts_path is not None and not self.artifacts_path.is_dir():
+            raise RuntimeError(
+                f"The value of {self.artifacts_path=} is not valid. "
+                "When defined, it must point to a folder containing all models required by the pipeline."
+            )
+
     def execute(self, in_doc: InputDocument, raises_on_error: bool) -> ConversionResult:
         conv_res = ConversionResult(input=in_doc)
 
@@ -108,15 +131,58 @@ class BasePipeline(ABC):
     def is_backend_supported(cls, backend: AbstractDocumentBackend):
         pass
 
-    # def _apply_on_elements(self, element_batch: Iterable[NodeItem]) -> Iterable[Any]:
-    #    for model in self.build_pipe:
-    #        element_batch = model(element_batch)
-    #
-    #    yield from element_batch
 
+class ConvertPipeline(BasePipeline):
+    def __init__(self, pipeline_options: ConvertPipelineOptions):
+        super().__init__(pipeline_options)
+        self.pipeline_options: ConvertPipelineOptions
 
-class PaginatedPipeline(BasePipeline):  # TODO this is a bad name.
-    def __init__(self, pipeline_options: PipelineOptions):
+        # ------ Common enrichment models working on all backends
+
+        # Picture description model
+        if (
+            picture_description_model := self._get_picture_description_model(
+                artifacts_path=self.artifacts_path
+            )
+        ) is None:
+            raise RuntimeError(
+                f"The specified picture description kind is not supported: {pipeline_options.picture_description_options.kind}."
+            )
+
+        self.enrichment_pipe = [
+            # Document Picture Classifier
+            DocumentPictureClassifier(
+                enabled=pipeline_options.do_picture_classification,
+                artifacts_path=self.artifacts_path,
+                options=DocumentPictureClassifierOptions(),
+                accelerator_options=pipeline_options.accelerator_options,
+            ),
+            # Document Picture description
+            picture_description_model,
+        ]
+
+    def _get_picture_description_model(
+        self, artifacts_path: Optional[Path] = None
+    ) -> Optional[PictureDescriptionBaseModel]:
+        factory = get_picture_description_factory(
+            allow_external_plugins=self.pipeline_options.allow_external_plugins
+        )
+        return factory.create_instance(
+            options=self.pipeline_options.picture_description_options,
+            enabled=self.pipeline_options.do_picture_description,
+            enable_remote_services=self.pipeline_options.enable_remote_services,
+            artifacts_path=artifacts_path,
+            accelerator_options=self.pipeline_options.accelerator_options,
+        )
+
+    @classmethod
+    @abstractmethod
+    def get_default_options(cls) -> ConvertPipelineOptions:
+        pass
+
+
+class PaginatedPipeline(ConvertPipeline):  # TODO this is a bad name.
+    def __init__(self, pipeline_options: ConvertPipelineOptions):
         super().__init__(pipeline_options)
         self.keep_backend = False
 

{docling-2.51.0 → docling-2.52.0}/docling/pipeline/extraction_vlm_pipeline.py

@@ -1,7 +1,6 @@
 import inspect
 import json
 import logging
-from pathlib import Path
 from typing import Optional
 
 from PIL.Image import Image
@@ -16,7 +15,10 @@ from docling.datamodel.extraction import (
     ExtractionResult,
     ExtractionTemplateType,
 )
-from docling.datamodel.pipeline_options import BaseOptions, VlmExtractionPipelineOptions
+from docling.datamodel.pipeline_options import (
+    PipelineOptions,
+    VlmExtractionPipelineOptions,
+)
 from docling.datamodel.settings import settings
 from docling.models.vlm_models_inline.nuextract_transformers_model import (
     NuExtractTransformersModel,
@@ -35,22 +37,10 @@ class ExtractionVlmPipeline(BaseExtractionPipeline):
         self.accelerator_options = pipeline_options.accelerator_options
         self.pipeline_options: VlmExtractionPipelineOptions
 
-        artifacts_path: Optional[Path] = None
-        if pipeline_options.artifacts_path is not None:
-            artifacts_path = Path(pipeline_options.artifacts_path).expanduser()
-        elif settings.artifacts_path is not None:
-            artifacts_path = Path(settings.artifacts_path).expanduser()
-
-        if artifacts_path is not None and not artifacts_path.is_dir():
-            raise RuntimeError(
-                f"The value of {artifacts_path=} is not valid. "
-                "When defined, it must point to a folder containing all models required by the pipeline."
-            )
-
         # Create VLM model instance
         self.vlm_model = NuExtractTransformersModel(
             enabled=True,
-            artifacts_path=artifacts_path,  # Will download automatically
+            artifacts_path=self.artifacts_path,  # Will download automatically
             accelerator_options=self.accelerator_options,
             vlm_options=pipeline_options.vlm_options,
         )
@@ -194,11 +184,14 @@ class ExtractionVlmPipeline(BaseExtractionPipeline):
             class ExtractionTemplateFactory(ModelFactory[template]):  # type: ignore
                 __use_examples__ = True  # prefer Field(examples=...) when present
                 __use_defaults__ = True  # use field defaults instead of random values
+                __check_model__ = (
+                    True  # setting the value to avoid deprecation warnings
+                )
 
             return ExtractionTemplateFactory.build().model_dump_json(indent=2)  # type: ignore
         else:
             raise ValueError(f"Unsupported template type: {type(template)}")
 
     @classmethod
-    def get_default_options(cls) -> BaseOptions:
+    def get_default_options(cls) -> PipelineOptions:
         return VlmExtractionPipelineOptions()

{docling-2.51.0 → docling-2.52.0}/docling/pipeline/simple_pipeline.py

@@ -6,21 +6,21 @@ from docling.backend.abstract_backend import (
 )
 from docling.datamodel.base_models import ConversionStatus
 from docling.datamodel.document import ConversionResult
-from docling.datamodel.pipeline_options import PipelineOptions
-from docling.pipeline.base_pipeline import BasePipeline
+from docling.datamodel.pipeline_options import ConvertPipelineOptions
+from docling.pipeline.base_pipeline import ConvertPipeline
 from docling.utils.profiling import ProfilingScope, TimeRecorder
 
 _log = logging.getLogger(__name__)
 
 
-class SimplePipeline(BasePipeline):
+class SimplePipeline(ConvertPipeline):
     """SimpleModelPipeline.
 
     This class is used at the moment for formats / backends
     which produce straight DoclingDocument output.
     """
 
-    def __init__(self, pipeline_options: PipelineOptions):
+    def __init__(self, pipeline_options: ConvertPipelineOptions):
         super().__init__(pipeline_options)
 
     def _build_document(self, conv_res: ConversionResult) -> ConversionResult:
@@ -47,8 +47,8 @@ class SimplePipeline(BasePipeline):
         return ConversionStatus.SUCCESS
 
     @classmethod
-    def get_default_options(cls) -> PipelineOptions:
-        return PipelineOptions()
+    def get_default_options(cls) -> ConvertPipelineOptions:
+        return ConvertPipelineOptions()
 
     @classmethod
     def is_backend_supported(cls, backend: AbstractDocumentBackend):