diff --git a/Dockerfile b/Dockerfile
index dd87b0fb..9bed78f3 100644
--- a/Dockerfile
+++ b/Dockerfile
@@ -17,7 +17,7 @@ ENV TORCH_HOME=/tmp/
 COPY examples/minimal.py /root/minimal.py
 
 RUN python -c 'from deepsearch_glm.utils.load_pretrained_models import load_pretrained_nlp_models; load_pretrained_nlp_models(verbose=True);'
-RUN python -c 'from docling.document_converter import DocumentConverter; artifacts_path = DocumentConverter.download_models_hf(force=True);'
+RUN python -c 'from docling.pipeline.standard_pdf_pipeline import StandardPdfPipeline; StandardPdfPipeline.download_models_hf(force=True);'
 
 # On container environments, always set a thread budget to avoid undesired thread congestion.
 ENV OMP_NUM_THREADS=4