Update packages

docs/linux_install.sh

@@ -53,9 +53,9 @@ sudo apt-get install -y libmagic-dev poppler-utils tesseract-ocr libtesseract-de
 # Optional: For DocTR
 pip install -r reqs_optional/requirements_optional_doctr.txt -c reqs_optional/reqs_constraints.txt
 # For DocTR: go back to older onnx so Tesseract OCR still works
-pip install onnxruntime==1.15.0 -c reqs_optional/reqs_constraints.txt
+pip install onnxruntime==1.16.0 -c reqs_optional/reqs_constraints.txt
 # GPU only:
-pip install onnxruntime-gpu==1.15.0 -c reqs_optional/reqs_constraints.txt
+pip install onnxruntime-gpu==1.16.0 -c reqs_optional/reqs_constraints.txt
 
 # Optional: for supporting unstructured package
 for i in 1 2 3 4; do python -m nltk.downloader all && break || sleep 1; done  # retry as frequently fails with github downloading issues

docs/windows_freezelist.txt

@@ -211,9 +211,9 @@ numpy==1.23.4
 oauthlib==3.2.2
 olefile==0.47
 omegaconf==2.3.0
-onnx==1.15.0
+onnx==1.16.0
 onnxruntime==1.15.0
-onnxruntime-gpu==1.15.0
+onnxruntime-gpu==1.16.0
 openai==1.3.7
 opencv-python==4.9.0.80
 opencv-python-headless==4.9.0.80
@@ -400,7 +400,7 @@ unicodedata2==15.1.0 # @ file:///D:/bld/unicodedata2_1695848155043/work
 Unidecode==1.3.7
 universal-analytics-python3==1.1.1
 unstructured==0.11.8
-unstructured-inference==0.7.15
+unstructured-inference==0.7.25
 unstructured.pytesseract==0.3.12
 uri-template==1.3.0
 urllib3==1.26.18

docs/windows_install.bat

@@ -36,9 +36,9 @@ REM     # Note: for Selenium, we match versions of playwright so above installer
 REM    # Optional: For DocTR
 CALL pip install -r reqs_optional/requirements_optional_doctr.txt -c reqs_optional/reqs_constraints.txt
 REM      # For DocTR: go back to older onnx so Tesseract OCR still works
-CALL pip install onnxruntime==1.15.0 -c reqs_optional/reqs_constraints.txt
+CALL pip install onnxruntime==1.16.0 -c reqs_optional/reqs_constraints.txt
 REM      # GPU only:
-CALL pip install onnxruntime-gpu==1.15.0 -c reqs_optional/reqs_constraints.txt
+CALL pip install onnxruntime-gpu==1.16.0 -c reqs_optional/reqs_constraints.txt
 
 REM # Audio transcription from Youtube videos and local mp3 files:
 REM Only for Microsoft TTS, not Coqui

reqs_optional/requirements_optional_cpu_only.txt

@@ -1,3 +1,3 @@
 faiss-cpu>=1.7.4
 # for unstructured
-onnxruntime==1.15.0
+onnxruntime==1.16.0

reqs_optional/requirements_optional_doctr.txt

@@ -1,4 +1,4 @@
-python-doctr @ git+https://github.com/h2oai/doctr.git@aee9b1c369e37af9e18265660935bce2c4447d65
+python-doctr @ git+https://github.com/h2oai/doctr.git@fce37e2f67e54a795820b55cd03f56e6982a424e
 weasyprint>=60.1
 imutils>=0.5.4
 opencv-python-headless>=4.8.1.78

reqs_optional/requirements_optional_gpu_only.txt

@@ -1,6 +1,6 @@
 faiss-gpu>=1.7.2
 # for unstructured
-onnxruntime-gpu==1.15.0
+onnxruntime-gpu==1.16.0
 auto-gptq==0.7.1
 optimum==1.18.0
 # autoawq for cuda 12.1, else build from source: https://github.com/casper-hansen/AutoAWQ?tab=readme-ov-file#build-from-source

reqs_optional/requirements_optional_langchain.txt

@@ -59,8 +59,9 @@ pydantic-settings==2.1.0
 
 # strong support for images
 # Requires on Ubuntu: sudo apt-get install libmagic-dev poppler-utils tesseract-ocr libtesseract-dev libreoffice
-unstructured[local-inference]==0.12.5
-unstructured[all-docs]==0.12.5
+unstructured[local-inference]==0.13.0
+unstructured[all-docs]==0.13.0
+unstructured-inference==0.7.25
 docx2txt==0.8
 python-docx==1.1.0
 #pdf2image==1.16.3

windows_installer.cfg

@@ -223,7 +223,7 @@ pypi_wheels = absl-py==2.0.0
     oauthlib==3.2.2
     olefile==0.47
     omegaconf==2.3.0
-    onnx==1.15.0
+    onnx==1.16.0
     onnxruntime==1.15.0
     onnxruntime-gpu==1.15.0
     openai==1.3.7
@@ -412,7 +412,7 @@ pypi_wheels = absl-py==2.0.0
     Unidecode==1.3.7
     universal-analytics-python3==1.1.1
     unstructured==0.11.8
-    unstructured-inference==0.7.15
+    unstructured-inference==0.7.25
     unstructured.pytesseract==0.3.12
     uri-template==1.3.0
     urllib3==1.26.18