mirrors/docling
1
0
Fork 0
mirror of https://github.com/DS4SD/docling.git synced 2025-12-08 12:48:28 +00:00
Code Issues Packages Projects Releases Wiki Activity

adding granite-docling preview

Browse Source 
Signed-off-by: Peter Staar <taa@zurich.ibm.com>
This commit is contained in:
Peter Staar
2025-09-05 15:00:05 +02:00
parent b49d1ad4f1
commit c1dcb0597d
2 changed files with 38 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

14
docling/cli/main.py
View File

@@ -63,6 +63,8 @@ from docling.datamodel.vlm_model_specs import (
GOT2_TRANSFORMERS,
GRANITE_VISION_OLLAMA,
GRANITE_VISION_TRANSFORMERS,
GRANITEDOCLING_MLX,
GRANITEDOCLING_TRANSFORMERS,
SMOLDOCLING_MLX,
SMOLDOCLING_TRANSFORMERS,
SMOLDOCLING_VLLM,
@@ -655,6 +657,18 @@ def convert( # noqa: C901
"To run SmolDocling faster, please install mlx-vlm:\n"
"pip install mlx-vlm"
)
elif vlm_model == VlmModelType.GRANITEDOCLING:
pipeline_options.vlm_options = GRANITEDOCLING_TRANSFORMERS
if sys.platform == "darwin":
try:
import mlx_vlm
pipeline_options.vlm_options = GRANITEDOCLING_MLX
except ImportError:
_log.warning(
"To run SmolDocling faster, please install mlx-vlm:\n"
"pip install mlx-vlm"
)
elif vlm_model == VlmModelType.SMOLDOCLING_VLLM:
pipeline_options.vlm_options = SMOLDOCLING_VLLM

24
docling/datamodel/vlm_model_specs.py
View File

@@ -18,6 +18,29 @@ from docling.datamodel.pipeline_options_vlm_model import (
_log = logging.getLogger(__name__)
# Granite-Docling
GRANITEDOCLING_TRANSFORMERS = InlineVlmOptions(
repo_id="ds4sd/granite-docling-258m-2-9-2025-v2",
prompt="Convert this page to docling.",
response_format=ResponseFormat.DOCTAGS,
inference_framework=InferenceFramework.MLX,
supported_devices=[AcceleratorDevice.MPS],
scale=2.0,
temperature=0.0,
stop_strings=["</doctag>", "<end_of_utterance>"],
)
GRANITEDOCLING_MLX = InlineVlmOptions(
repo_id="ds4sd/granite-docling-258m-2-9-2025-v2-mlx-bf16",
prompt="Convert this page to docling.",
response_format=ResponseFormat.DOCTAGS,
inference_framework=InferenceFramework.MLX,
supported_devices=[AcceleratorDevice.MPS],
scale=2.0,
temperature=0.0,
stop_strings=["</doctag>", "<end_of_utterance>"],
)
# SmolDocling
SMOLDOCLING_MLX = InlineVlmOptions(
repo_id="ds4sd/SmolDocling-256M-preview-mlx-bf16",
@@ -272,3 +295,4 @@ class VlmModelType(str, Enum):
GRANITE_VISION_VLLM = "granite_vision_vllm"
GRANITE_VISION_OLLAMA = "granite_vision_ollama"
GOT_OCR_2 = "got_ocr_2"
GRANITEDOCLING = "granite_docling"