docling 2.51.0__py3-none-any.whl → 2.53.0__py3-none-any.whl

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/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):

docling/pipeline/standard_pdf_pipeline.py

@@ -15,18 +15,13 @@ from docling.datamodel.pipeline_options import PdfPipelineOptions
 from docling.datamodel.settings import settings
 from docling.models.base_ocr_model import BaseOcrModel
 from docling.models.code_formula_model import CodeFormulaModel, CodeFormulaModelOptions
-from docling.models.document_picture_classifier import (
-    DocumentPictureClassifier,
-    DocumentPictureClassifierOptions,
-)
-from docling.models.factories import get_ocr_factory, get_picture_description_factory
+from docling.models.factories import get_ocr_factory
 from docling.models.layout_model import LayoutModel
 from docling.models.page_assemble_model import PageAssembleModel, PageAssembleOptions
 from docling.models.page_preprocessing_model import (
     PagePreprocessingModel,
     PagePreprocessingOptions,
 )
-from docling.models.picture_description_base_model import PictureDescriptionBaseModel
 from docling.models.readingorder_model import ReadingOrderModel, ReadingOrderOptions
 from docling.models.table_structure_model import TableStructureModel
 from docling.pipeline.base_pipeline import PaginatedPipeline
@@ -41,18 +36,6 @@ class StandardPdfPipeline(PaginatedPipeline):
         super().__init__(pipeline_options)
         self.pipeline_options: PdfPipelineOptions
 
-        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."
-            )
-
         with warnings.catch_warnings():  # deprecated generate_table_images
             warnings.filterwarnings("ignore", category=DeprecationWarning)
             self.keep_images = (
@@ -63,7 +46,7 @@ class StandardPdfPipeline(PaginatedPipeline):
 
         self.reading_order_model = ReadingOrderModel(options=ReadingOrderOptions())
 
-        ocr_model = self.get_ocr_model(artifacts_path=artifacts_path)
+        ocr_model = self.get_ocr_model(artifacts_path=self.artifacts_path)
 
         self.build_pipe = [
             # Pre-processing
@@ -76,14 +59,14 @@ class StandardPdfPipeline(PaginatedPipeline):
             ocr_model,
             # Layout model
             LayoutModel(
-                artifacts_path=artifacts_path,
+                artifacts_path=self.artifacts_path,
                 accelerator_options=pipeline_options.accelerator_options,
                 options=pipeline_options.layout_options,
             ),
             # Table structure model
             TableStructureModel(
                 enabled=pipeline_options.do_table_structure,
-                artifacts_path=artifacts_path,
+                artifacts_path=self.artifacts_path,
                 options=pipeline_options.table_structure_options,
                 accelerator_options=pipeline_options.accelerator_options,
             ),
@@ -91,37 +74,19 @@ class StandardPdfPipeline(PaginatedPipeline):
             PageAssembleModel(options=PageAssembleOptions()),
         ]
 
-        # Picture description model
-        if (
-            picture_description_model := self.get_picture_description_model(
-                artifacts_path=artifacts_path
-            )
-        ) is None:
-            raise RuntimeError(
-                f"The specified picture description kind is not supported: {pipeline_options.picture_description_options.kind}."
-            )
-
         self.enrichment_pipe = [
             # Code Formula Enrichment Model
             CodeFormulaModel(
                 enabled=pipeline_options.do_code_enrichment
                 or pipeline_options.do_formula_enrichment,
-                artifacts_path=artifacts_path,
+                artifacts_path=self.artifacts_path,
                 options=CodeFormulaModelOptions(
                     do_code_enrichment=pipeline_options.do_code_enrichment,
                     do_formula_enrichment=pipeline_options.do_formula_enrichment,
                 ),
                 accelerator_options=pipeline_options.accelerator_options,
             ),
-            # Document Picture Classifier
-            DocumentPictureClassifier(
-                enabled=pipeline_options.do_picture_classification,
-                artifacts_path=artifacts_path,
-                options=DocumentPictureClassifierOptions(),
-                accelerator_options=pipeline_options.accelerator_options,
-            ),
-            # Document Picture description
-            picture_description_model,
+            *self.enrichment_pipe,
         ]
 
         if (
@@ -158,20 +123,6 @@ class StandardPdfPipeline(PaginatedPipeline):
             accelerator_options=self.pipeline_options.accelerator_options,
         )
 
-    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,
-        )
-
     def initialize_page(self, conv_res: ConversionResult, page: Page) -> Page:
         with TimeRecorder(conv_res, "page_init"):
             page._backend = conv_res.input._backend.load_page(page.page_no)  # type: ignore

docling/pipeline/threaded_standard_pdf_pipeline.py

@@ -20,10 +20,14 @@ import itertools
 import logging
 import threading
 import time
+import warnings
 from collections import defaultdict, deque
 from dataclasses import dataclass, field
 from pathlib import Path
-from typing import Any, Iterable, List, Optional, Sequence, Tuple
+from typing import Any, Iterable, List, Optional, Sequence, Tuple, cast
+
+import numpy as np
+from docling_core.types.doc import DocItem, ImageRef, PictureItem, TableItem
 
 from docling.backend.abstract_backend import AbstractDocumentBackend
 from docling.backend.pdf_backend import PdfDocumentBackend
@@ -32,21 +36,16 @@ from docling.datamodel.document import ConversionResult
 from docling.datamodel.pipeline_options import ThreadedPdfPipelineOptions
 from docling.datamodel.settings import settings
 from docling.models.code_formula_model import CodeFormulaModel, CodeFormulaModelOptions
-from docling.models.document_picture_classifier import (
-    DocumentPictureClassifier,
-    DocumentPictureClassifierOptions,
-)
-from docling.models.factories import get_ocr_factory, get_picture_description_factory
+from docling.models.factories import get_ocr_factory
 from docling.models.layout_model import LayoutModel
 from docling.models.page_assemble_model import PageAssembleModel, PageAssembleOptions
 from docling.models.page_preprocessing_model import (
     PagePreprocessingModel,
     PagePreprocessingOptions,
 )
-from docling.models.picture_description_base_model import PictureDescriptionBaseModel
 from docling.models.readingorder_model import ReadingOrderModel, ReadingOrderOptions
 from docling.models.table_structure_model import TableStructureModel
-from docling.pipeline.base_pipeline import BasePipeline
+from docling.pipeline.base_pipeline import ConvertPipeline
 from docling.utils.profiling import ProfilingScope, TimeRecorder
 from docling.utils.utils import chunkify
 
@@ -294,7 +293,7 @@ class RunContext:
 # ──────────────────────────────────────────────────────────────────────────────
 
 
-class ThreadedStandardPdfPipeline(BasePipeline):
+class ThreadedStandardPdfPipeline(ConvertPipeline):
     """High-performance PDF pipeline with multi-threaded stages."""
 
     def __init__(self, pipeline_options: ThreadedPdfPipelineOptions) -> None:
@@ -310,7 +309,7 @@ class ThreadedStandardPdfPipeline(BasePipeline):
     # ────────────────────────────────────────────────────────────────────────
 
     def _init_models(self) -> None:
-        art_path = self._resolve_artifacts_path()
+        art_path = self.artifacts_path
         self.keep_images = (
             self.pipeline_options.generate_page_images
             or self.pipeline_options.generate_picture_images
@@ -337,32 +336,20 @@ class ThreadedStandardPdfPipeline(BasePipeline):
         self.reading_order_model = ReadingOrderModel(options=ReadingOrderOptions())
 
         # --- optional enrichment ------------------------------------------------
-        self.enrichment_pipe = []
-        code_formula = CodeFormulaModel(
-            enabled=self.pipeline_options.do_code_enrichment
-            or self.pipeline_options.do_formula_enrichment,
-            artifacts_path=art_path,
-            options=CodeFormulaModelOptions(
-                do_code_enrichment=self.pipeline_options.do_code_enrichment,
-                do_formula_enrichment=self.pipeline_options.do_formula_enrichment,
+        self.enrichment_pipe = [
+            # Code Formula Enrichment Model
+            CodeFormulaModel(
+                enabled=self.pipeline_options.do_code_enrichment
+                or self.pipeline_options.do_formula_enrichment,
+                artifacts_path=self.artifacts_path,
+                options=CodeFormulaModelOptions(
+                    do_code_enrichment=self.pipeline_options.do_code_enrichment,
+                    do_formula_enrichment=self.pipeline_options.do_formula_enrichment,
+                ),
+                accelerator_options=self.pipeline_options.accelerator_options,
             ),
-            accelerator_options=self.pipeline_options.accelerator_options,
-        )
-        if code_formula.enabled:
-            self.enrichment_pipe.append(code_formula)
-
-        picture_classifier = DocumentPictureClassifier(
-            enabled=self.pipeline_options.do_picture_classification,
-            artifacts_path=art_path,
-            options=DocumentPictureClassifierOptions(),
-            accelerator_options=self.pipeline_options.accelerator_options,
-        )
-        if picture_classifier.enabled:
-            self.enrichment_pipe.append(picture_classifier)
-
-        picture_descr = self._make_picture_description_model(art_path)
-        if picture_descr and picture_descr.enabled:
-            self.enrichment_pipe.append(picture_descr)
+            *self.enrichment_pipe,
+        ]
 
         self.keep_backend = any(
             (
@@ -374,19 +361,6 @@ class ThreadedStandardPdfPipeline(BasePipeline):
         )
 
     # ---------------------------------------------------------------- helpers
-    def _resolve_artifacts_path(self) -> Optional[Path]:
-        if self.pipeline_options.artifacts_path:
-            p = Path(self.pipeline_options.artifacts_path).expanduser()
-        elif settings.artifacts_path:
-            p = Path(settings.artifacts_path).expanduser()
-        else:
-            return None
-        if not p.is_dir():
-            raise RuntimeError(
-                f"{p} does not exist or is not a directory containing the required models"
-            )
-        return p
-
     def _make_ocr_model(self, art_path: Optional[Path]) -> Any:
         factory = get_ocr_factory(
             allow_external_plugins=self.pipeline_options.allow_external_plugins
@@ -398,20 +372,6 @@ class ThreadedStandardPdfPipeline(BasePipeline):
             accelerator_options=self.pipeline_options.accelerator_options,
         )
 
-    def _make_picture_description_model(
-        self, art_path: Optional[Path]
-    ) -> 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=art_path,
-            accelerator_options=self.pipeline_options.accelerator_options,
-        )
-
     # ────────────────────────────────────────────────────────────────────────
     # Build - thread pipeline
     # ────────────────────────────────────────────────────────────────────────
@@ -585,6 +545,86 @@ class ThreadedStandardPdfPipeline(BasePipeline):
                 elements=elements, headers=headers, body=body
             )
             conv_res.document = self.reading_order_model(conv_res)
+
+            # Generate page images in the output
+            if self.pipeline_options.generate_page_images:
+                for page in conv_res.pages:
+                    assert page.image is not None
+                    page_no = page.page_no + 1
+                    conv_res.document.pages[page_no].image = ImageRef.from_pil(
+                        page.image, dpi=int(72 * self.pipeline_options.images_scale)
+                    )
+
+            # Generate images of the requested element types
+            with warnings.catch_warnings():  # deprecated generate_table_images
+                warnings.filterwarnings("ignore", category=DeprecationWarning)
+                if (
+                    self.pipeline_options.generate_picture_images
+                    or self.pipeline_options.generate_table_images
+                ):
+                    scale = self.pipeline_options.images_scale
+                    for element, _level in conv_res.document.iterate_items():
+                        if not isinstance(element, DocItem) or len(element.prov) == 0:
+                            continue
+                        if (
+                            isinstance(element, PictureItem)
+                            and self.pipeline_options.generate_picture_images
+                        ) or (
+                            isinstance(element, TableItem)
+                            and self.pipeline_options.generate_table_images
+                        ):
+                            page_ix = element.prov[0].page_no - 1
+                            page = next(
+                                (p for p in conv_res.pages if p.page_no == page_ix),
+                                cast("Page", None),
+                            )
+                            assert page is not None
+                            assert page.size is not None
+                            assert page.image is not None
+
+                            crop_bbox = (
+                                element.prov[0]
+                                .bbox.scaled(scale=scale)
+                                .to_top_left_origin(
+                                    page_height=page.size.height * scale
+                                )
+                            )
+
+                            cropped_im = page.image.crop(crop_bbox.as_tuple())
+                            element.image = ImageRef.from_pil(
+                                cropped_im, dpi=int(72 * scale)
+                            )
+
+            # Aggregate confidence values for document:
+            if len(conv_res.pages) > 0:
+                with warnings.catch_warnings():
+                    warnings.filterwarnings(
+                        "ignore",
+                        category=RuntimeWarning,
+                        message="Mean of empty slice|All-NaN slice encountered",
+                    )
+                    conv_res.confidence.layout_score = float(
+                        np.nanmean(
+                            [c.layout_score for c in conv_res.confidence.pages.values()]
+                        )
+                    )
+                    conv_res.confidence.parse_score = float(
+                        np.nanquantile(
+                            [c.parse_score for c in conv_res.confidence.pages.values()],
+                            q=0.1,  # parse score should relate to worst 10% of pages.
+                        )
+                    )
+                    conv_res.confidence.table_score = float(
+                        np.nanmean(
+                            [c.table_score for c in conv_res.confidence.pages.values()]
+                        )
+                    )
+                    conv_res.confidence.ocr_score = float(
+                        np.nanmean(
+                            [c.ocr_score for c in conv_res.confidence.pages.values()]
+                        )
+                    )
+
         return conv_res
 
     # ---------------------------------------------------------------- misc

docling/pipeline/vlm_pipeline.py

@@ -54,18 +54,6 @@ class VlmPipeline(PaginatedPipeline):
 
         self.pipeline_options: VlmPipelineOptions
 
-        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."
-            )
-
         # force_backend_text = False - use text that is coming from VLM response
         # force_backend_text = True - get text from backend using bounding boxes predicted by SmolDocling doctags
         self.force_backend_text = (
@@ -89,7 +77,7 @@ class VlmPipeline(PaginatedPipeline):
                 self.build_pipe = [
                     HuggingFaceMlxModel(
                         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,
                         vlm_options=vlm_options,
                     ),
@@ -98,7 +86,7 @@ class VlmPipeline(PaginatedPipeline):
                 self.build_pipe = [
                     HuggingFaceTransformersVlmModel(
                         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,
                         vlm_options=vlm_options,
                     ),
@@ -109,7 +97,7 @@ class VlmPipeline(PaginatedPipeline):
                 self.build_pipe = [
                     VllmVlmModel(
                         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,
                         vlm_options=vlm_options,
                     ),

docling/utils/model_downloader.py

@@ -10,6 +10,8 @@ from docling.datamodel.pipeline_options import (
 )
 from docling.datamodel.settings import settings
 from docling.datamodel.vlm_model_specs import (
+    GRANITEDOCLING_MLX,
+    GRANITEDOCLING_TRANSFORMERS,
     SMOLDOCLING_MLX,
     SMOLDOCLING_TRANSFORMERS,
 )
@@ -34,6 +36,8 @@ def download_models(
     with_code_formula: bool = True,
     with_picture_classifier: bool = True,
     with_smolvlm: bool = False,
+    with_granitedocling: bool = False,
+    with_granitedocling_mlx: bool = False,
     with_smoldocling: bool = False,
     with_smoldocling_mlx: bool = False,
     with_granite_vision: bool = False,
@@ -86,6 +90,24 @@ def download_models(
             progress=progress,
         )
 
+    if with_granitedocling:
+        _log.info("Downloading GraniteDocling model...")
+        download_hf_model(
+            repo_id=GRANITEDOCLING_TRANSFORMERS.repo_id,
+            local_dir=output_dir / GRANITEDOCLING_TRANSFORMERS.repo_cache_folder,
+            force=force,
+            progress=progress,
+        )
+
+    if with_granitedocling_mlx:
+        _log.info("Downloading GraniteDocling MLX model...")
+        download_hf_model(
+            repo_id=GRANITEDOCLING_MLX.repo_id,
+            local_dir=output_dir / GRANITEDOCLING_MLX.repo_cache_folder,
+            force=force,
+            progress=progress,
+        )
+
     if with_smoldocling:
         _log.info("Downloading SmolDocling model...")
         download_hf_model(

{docling-2.51.0.dist-info → docling-2.53.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: docling
-Version: 2.51.0
+Version: 2.53.0
 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
@@ -108,18 +108,22 @@ Docling simplifies document processing, parsing diverse formats — including ad
 * 🔒 Local execution capabilities for sensitive data and air-gapped environments
 * 🤖 Plug-and-play [integrations][integrations] incl. LangChain, LlamaIndex, Crew AI & Haystack for agentic AI
 * 🔍 Extensive OCR support for scanned PDFs and images
-* 👓 Support of several Visual Language Models ([SmolDocling](https://huggingface.co/ds4sd/SmolDocling-256M-preview))
+* 👓 Support of several Visual Language Models ([GraniteDocling](https://huggingface.co/ibm-granite/granite-docling-258M))
 * 🎙️ 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
@@ -156,9 +160,9 @@ Docling has a built-in CLI to run conversions.
 docling https://arxiv.org/pdf/2206.01062
 ```
 
-You can also use 🥚[SmolDocling](https://huggingface.co/ds4sd/SmolDocling-256M-preview) and other VLMs via Docling CLI:
+You can also use 🥚[GraniteDocling](https://huggingface.co/ibm-granite/granite-docling-258M) and other VLMs via Docling CLI:
 ```bash
-docling --pipeline vlm --vlm-model smoldocling https://arxiv.org/pdf/2206.01062
+docling --pipeline vlm --vlm-model granite_docling https://arxiv.org/pdf/2206.01062
 ```
 This will use MLX acceleration on supported Apple Silicon hardware.