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

Small fixes

Browse Source 
Signed-off-by: Christoph Auer <cau@zurich.ibm.com>
This commit is contained in:
Christoph Auer
2025-08-18 11:12:02 +02:00
parent f42676aab9
commit e372cfe01a
2 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
docling/datamodel/vlm_model_specs.py
View File

@@ -65,7 +65,7 @@ SMOLDOCLING_VLLM = InlineVlmOptions(
SMOLVLM500_TRANSFORMERS = InlineVlmOptions( SMOLVLM500_TRANSFORMERS = InlineVlmOptions(
repo_id="HuggingFaceTB/SmolVLM-500M-Instruct", repo_id="HuggingFaceTB/SmolVLM-500M-Instruct",
prompt="Transcribe this image to plain text.", prompt="Transcribe this image to plain text.",
response_format=ResponseFormat.DOCTAGS, response_format=ResponseFormat.PLAINTEXT,
inference_framework=InferenceFramework.TRANSFORMERS, inference_framework=InferenceFramework.TRANSFORMERS,
transformers_model_type=TransformersModelType.AUTOMODEL_IMAGETEXTTOTEXT, transformers_model_type=TransformersModelType.AUTOMODEL_IMAGETEXTTOTEXT,
supported_devices=[ supported_devices=[

2
docling/pipeline/threaded_multistage_vlm_pipeline.py
View File

@@ -161,7 +161,7 @@ class ThreadedMultiStageVlmPipelineOptions(PaginatedPipelineOptions):
# text_opts = DOLPHIN_TRANSFORMERS.model_copy() # text_opts = DOLPHIN_TRANSFORMERS.model_copy()
# text_opts.prompt = "<s>Read text in the image. <Answer/>" # text_opts.prompt = "<s>Read text in the image. <Answer/>"
base_model = SMOLVLM500_TRANSFORMERS base_model = SMOLVLM500_MLX
text_opts = base_model.model_copy() text_opts = base_model.model_copy()
# text_opts.prompt = "Convert this page to docling." # text_opts.prompt = "Convert this page to docling."