LayoutDetection下任何PP-DocLayout-xxx模型，在CPU高性能推理下以openvino或onnxruntime作为推理后端时，无法解析多页PDF，且报错退出 #4615
Description
Checklist:
- 查找历史相关issue寻求解答
- 翻阅FAQ
- 翻阅PaddleX 文档
- 确认bug是否在新版本里还未修复
描述问题
在CPU高性能推理模式下部署“通用版面解析v3产线”时，PP-DocLayout-M模型推理后端可使用openvino或者onnxruntime；但如果此时传入多页PDF，会报错退出。
如果改用paddle推理后端则能正常解析多页PDF。
另一个正常情形是将多页PDF拆分成单页PDF，然后逐页传入。但该方案违反了接口约定，不能使用。
复现
-
高性能推理
- 您是否完全按照高性能推理文档教程跑通了流程？
【是】
-
服务化部署
- 您是否完全按照服务化部署文档教程跑通了流程？
【是】
* 您在服务化部署中是否有使用高性能推理插件？
【是】
* 您使用了哪一种服务化部署方案？
docker自行打包部署（从飞浆3.0.0镜像直接构建），CPU高性能推理模式。
其中Dockerfile主要内容简化如下：
FROM ccr-2vdh3abv-pub.cnc.bj.baidubce.com/paddlepaddle/paddle:3.0.0 AS build
RUN <<-DOCKERFILERUNEOF
set -xe
python -m pip install "onnxruntime==1.22.0"
python -m pip install onnx_graphsurgeon==0.5.8
python -m pip install "paddlex[ocr]==3.2.1"
paddlex --install paddle2onnx
paddlex --install hpi-cpu
paddlex --install serving
DOCKERFILERUNEOF
- 您使用的模型和数据集是？
模型：
PP-StructureV3-debug.yaml配置如下（使用paddlex命令行直接导出产线文件，仅修改PP-DocLayout_plus-L为PP-DocLayout-M）：
pipeline_name: PP-StructureV3
batch_size: 8
use_doc_preprocessor: False
use_seal_recognition: False
use_table_recognition: True
use_formula_recognition: True
use_chart_recognition: False
use_region_detection: True
SubModules:
LayoutDetection:
module_name: layout_detection
model_name: PP-DocLayout-M
model_dir: null
batch_size: 8
threshold:
0: 0.3 # paragraph_title
1: 0.5 # image
2: 0.4 # text
3: 0.5 # number
4: 0.5 # abstract
5: 0.5 # content
6: 0.5 # figure_table_chart_title
7: 0.3 # formula
8: 0.5 # table
9: 0.5 # reference
10: 0.5 # doc_title
11: 0.5 # footnote
12: 0.5 # header
13: 0.5 # algorithm
14: 0.5 # footer
15: 0.45 # seal
16: 0.5 # chart
17: 0.5 # formula_number
18: 0.5 # aside_text
19: 0.5 # reference_content
layout_nms: True
layout_unclip_ratio: [1.0, 1.0]
layout_merge_bboxes_mode:
0: "large" # paragraph_title
1: "large" # image
2: "union" # text
3: "union" # number
4: "union" # abstract
5: "union" # content
6: "union" # figure_table_chart_title
7: "large" # formula
8: "union" # table
9: "union" # reference
10: "union" # doc_title
11: "union" # footnote
12: "union" # header
13: "union" # algorithm
14: "union" # footer
15: "union" # seal
16: "large" # chart
17: "union" # formula_number
18: "union" # aside_text
19: "union" # reference_content
ChartRecognition:
module_name: chart_recognition
model_name: PP-Chart2Table
model_dir: null
batch_size: 1
RegionDetection:
module_name: layout_detection
model_name: PP-DocBlockLayout
model_dir: null
layout_nms: True
layout_merge_bboxes_mode: "small"
SubPipelines:
DocPreprocessor:
pipeline_name: doc_preprocessor
batch_size: 8
use_doc_orientation_classify: True
use_doc_unwarping: True
SubModules:
DocOrientationClassify:
module_name: doc_text_orientation
model_name: PP-LCNet_x1_0_doc_ori
model_dir: null
batch_size: 8
DocUnwarping:
module_name: image_unwarping
model_name: UVDoc
model_dir: null
GeneralOCR:
pipeline_name: OCR
batch_size: 8
text_type: general
use_doc_preprocessor: False
use_textline_orientation: True
SubModules:
TextDetection:
module_name: text_detection
model_name: PP-OCRv5_server_det
model_dir: null
limit_side_len: 736
limit_type: min
max_side_limit: 4000
thresh: 0.3
box_thresh: 0.6
unclip_ratio: 1.5
TextLineOrientation:
module_name: textline_orientation
model_name: PP-LCNet_x1_0_textline_ori
model_dir: null
batch_size: 8
TextRecognition:
module_name: text_recognition
model_name: PP-OCRv5_server_rec
model_dir: null
batch_size: 8
score_thresh: 0.0
TableRecognition:
pipeline_name: table_recognition_v2
use_layout_detection: False
use_doc_preprocessor: False
use_ocr_model: False
SubModules:
TableClassification:
module_name: table_classification
model_name: PP-LCNet_x1_0_table_cls
model_dir: null
WiredTableStructureRecognition:
module_name: table_structure_recognition
model_name: SLANeXt_wired
model_dir: null
WirelessTableStructureRecognition:
module_name: table_structure_recognition
model_name: SLANet_plus
model_dir: null
WiredTableCellsDetection:
module_name: table_cells_detection
model_name: RT-DETR-L_wired_table_cell_det
model_dir: null
WirelessTableCellsDetection:
module_name: table_cells_detection
model_name: RT-DETR-L_wireless_table_cell_det
model_dir: null
TableOrientationClassify:
module_name: doc_text_orientation
model_name: PP-LCNet_x1_0_doc_ori
model_dir: null
SubPipelines:
GeneralOCR:
pipeline_name: OCR
text_type: general
use_doc_preprocessor: False
use_textline_orientation: True
SubModules:
TextDetection:
module_name: text_detection
model_name: PP-OCRv5_server_det
model_dir: null
limit_side_len: 736
limit_type: min
max_side_limit: 4000
thresh: 0.3
box_thresh: 0.4
unclip_ratio: 1.5
TextLineOrientation:
module_name: textline_orientation
model_name: PP-LCNet_x1_0_textline_ori
model_dir: null
batch_size: 8
TextRecognition:
module_name: text_recognition
model_name: PP-OCRv5_server_rec
model_dir: null
batch_size: 8
score_thresh: 0.0
SealRecognition:
pipeline_name: seal_recognition
batch_size: 8
use_layout_detection: False
use_doc_preprocessor: False
SubPipelines:
SealOCR:
pipeline_name: OCR
batch_size: 8
text_type: seal
use_doc_preprocessor: False
use_textline_orientation: False
SubModules:
TextDetection:
module_name: seal_text_detection
model_name: PP-OCRv4_server_seal_det
model_dir: null
limit_side_len: 736
limit_type: min
max_side_limit: 4000
thresh: 0.2
box_thresh: 0.6
unclip_ratio: 0.5
TextRecognition:
module_name: text_recognition
model_name: PP-OCRv5_server_rec
model_dir: null
batch_size: 8
score_thresh: 0
FormulaRecognition:
pipeline_name: formula_recognition
batch_size: 8
use_layout_detection: False
use_doc_preprocessor: False
SubModules:
FormulaRecognition:
module_name: formula_recognition
model_name: PP-FormulaNet_plus-L
model_dir: null
batch_size: 8
数据集：
任意多页PDF文件。
- 请提供您出现的报错信息及相关log
运行cli，其中1111111.pdf为多页PDF：
paddlex --pipeline /paddle/PP-StructureV3-debug.yaml --input /paddle/1111111.pdf --use_doc_orientation_classify true --use_doc_unwarping true --use_textline_orientation true --use_general_ocr true --use_table_recognition true --use_formula_recognition True --use_e2e_wired_table_rec_model false --use_e2e_wireless_table_rec_model True --save_path /paddle/outputdebug --device cpu --use_hpip
当PP-DocLayout-M模型推理后端使用openvino时：
Creating model: ('PP-LCNet_x1_0_doc_ori', None)
Model files already exist. Using cached files. To redownload, please delete the directory manually: `/.paddlex/official_models/PP-LCNet_x1_0_doc_ori`.
grep: warning: GREP_OPTIONS is deprecated; please use an alias or script
Inference backend: openvino
Inference backend config: cpu_num_threads=10
[INFO] ultra_infer/runtime/backends/openvino/ov_backend.cc(371)::InitFromOnnx number of streams:1.
[INFO] ultra_infer/runtime/backends/openvino/ov_backend.cc(375)::InitFromOnnx affinity:YES.
[INFO] ultra_infer/runtime/backends/openvino/ov_backend.cc(387)::InitFromOnnx Compile OpenVINO model on device_name:CPU.
[INFO] ultra_infer/runtime/runtime.cc(283)::CreateOpenVINOBackend Runtime initialized with Backend::OPENVINO in Device::CPU.
Creating model: ('UVDoc', None)
Model files already exist. Using cached files. To redownload, please delete the directory manually: `/.paddlex/official_models/UVDoc`.
The Paddle Inference backend is selected with the default configuration. This may not provide optimal performance.
Using Paddle Inference backend
Paddle predictor option: device_type: cpu, device_id: None, run_mode: paddle, trt_dynamic_shapes: {'img': [[1, 3, 128, 64], [1, 3, 256, 128], [8, 3, 512, 256]]}, cpu_threads: 10, delete_pass: [], enable_new_ir: True, enable_cinn: False, trt_cfg_setting: {}, trt_use_dynamic_shapes: True, trt_collect_shape_range_info: True, trt_discard_cached_shape_range_info: False, trt_dynamic_shape_input_data: None, trt_shape_range_info_path: None, trt_allow_rebuild_at_runtime: True, mkldnn_cache_capacity: 10
Creating model: ('PP-DocBlockLayout', None)
Model files already exist. Using cached files. To redownload, please delete the directory manually: `/.paddlex/official_models/PP-DocBlockLayout`.
Inference backend: onnxruntime
Inference backend config: cpu_num_threads=10
[INFO] ultra_infer/runtime/runtime.cc(308)::CreateOrtBackend Runtime initialized with Backend::ORT in Device::CPU.
Creating model: ('PP-DocLayout-M', None)
Model files already exist. Using cached files. To redownload, please delete the directory manually: `/.paddlex/official_models/PP-DocLayout-M`.
Inference backend: openvino
Inference backend config: cpu_num_threads=10
[INFO] ultra_infer/runtime/backends/openvino/ov_backend.cc(371)::InitFromOnnx number of streams:1.
[INFO] ultra_infer/runtime/backends/openvino/ov_backend.cc(375)::InitFromOnnx affinity:YES.
[INFO] ultra_infer/runtime/backends/openvino/ov_backend.cc(387)::InitFromOnnx Compile OpenVINO model on device_name:CPU.
[INFO] ultra_infer/runtime/runtime.cc(283)::CreateOpenVINOBackend Runtime initialized with Backend::OPENVINO in Device::CPU.
Creating model: ('PP-LCNet_x1_0_textline_ori', None)
Model files already exist. Using cached files. To redownload, please delete the directory manually: `/.paddlex/official_models/PP-LCNet_x1_0_textline_ori`.
The Paddle Inference backend is selected with the default configuration. This may not provide optimal performance.
Using Paddle Inference backend
Paddle predictor option: device_type: cpu, device_id: None, run_mode: mkldnn, trt_dynamic_shapes: {'x': [[1, 3, 80, 160], [1, 3, 80, 160], [8, 3, 80, 160]]}, cpu_threads: 10, delete_pass: [], enable_new_ir: True, enable_cinn: False, trt_cfg_setting: {}, trt_use_dynamic_shapes: True, trt_collect_shape_range_info: True, trt_discard_cached_shape_range_info: False, trt_dynamic_shape_input_data: None, trt_shape_range_info_path: None, trt_allow_rebuild_at_runtime: True, mkldnn_cache_capacity: 10
Creating model: ('PP-OCRv5_server_det', None)
Model files already exist. Using cached files. To redownload, please delete the directory manually: `/.paddlex/official_models/PP-OCRv5_server_det`.
The Paddle Inference backend is selected with the default configuration. This may not provide optimal performance.
Using Paddle Inference backend
Paddle predictor option: device_type: cpu, device_id: None, run_mode: mkldnn, trt_dynamic_shapes: {'x': [[1, 3, 32, 32], [1, 3, 736, 736], [1, 3, 4000, 4000]]}, cpu_threads: 10, delete_pass: [], enable_new_ir: True, enable_cinn: False, trt_cfg_setting: {}, trt_use_dynamic_shapes: True, trt_collect_shape_range_info: True, trt_discard_cached_shape_range_info: False, trt_dynamic_shape_input_data: None, trt_shape_range_info_path: None, trt_allow_rebuild_at_runtime: True, mkldnn_cache_capacity: 10
Creating model: ('PP-OCRv5_server_rec', None)
Model files already exist. Using cached files. To redownload, please delete the directory manually: `/.paddlex/official_models/PP-OCRv5_server_rec`.
The Paddle Inference backend is selected with the default configuration. This may not provide optimal performance.
Using Paddle Inference backend
Paddle predictor option: device_type: cpu, device_id: None, run_mode: mkldnn, trt_dynamic_shapes: {'x': [[1, 3, 48, 160], [1, 3, 48, 320], [8, 3, 48, 3200]]}, cpu_threads: 10, delete_pass: [], enable_new_ir: True, enable_cinn: False, trt_cfg_setting: {}, trt_use_dynamic_shapes: True, trt_collect_shape_range_info: True, trt_discard_cached_shape_range_info: False, trt_dynamic_shape_input_data: None, trt_shape_range_info_path: None, trt_allow_rebuild_at_runtime: True, mkldnn_cache_capacity: 10
Creating model: ('PP-LCNet_x1_0_table_cls', None)
Model files already exist. Using cached files. To redownload, please delete the directory manually: `/.paddlex/official_models/PP-LCNet_x1_0_table_cls`.
Inference backend: openvino
Inference backend config: cpu_num_threads=10
[INFO] ultra_infer/runtime/backends/openvino/ov_backend.cc(371)::InitFromOnnx number of streams:1.
[INFO] ultra_infer/runtime/backends/openvino/ov_backend.cc(375)::InitFromOnnx affinity:YES.
[INFO] ultra_infer/runtime/backends/openvino/ov_backend.cc(387)::InitFromOnnx Compile OpenVINO model on device_name:CPU.
[INFO] ultra_infer/runtime/runtime.cc(283)::CreateOpenVINOBackend Runtime initialized with Backend::OPENVINO in Device::CPU.
Creating model: ('SLANeXt_wired', None)
Model files already exist. Using cached files. To redownload, please delete the directory manually: `/.paddlex/official_models/SLANeXt_wired`.
Inference backend: onnxruntime
Inference backend config: cpu_num_threads=10
[INFO] ultra_infer/runtime/runtime.cc(308)::CreateOrtBackend Runtime initialized with Backend::ORT in Device::CPU.
Creating model: ('SLANet_plus', None)
Model files already exist. Using cached files. To redownload, please delete the directory manually: `/.paddlex/official_models/SLANet_plus`.
Inference backend: onnxruntime
Inference backend config: cpu_num_threads=10
[INFO] ultra_infer/runtime/runtime.cc(308)::CreateOrtBackend Runtime initialized with Backend::ORT in Device::CPU.
Creating model: ('RT-DETR-L_wired_table_cell_det', None)
Model files already exist. Using cached files. To redownload, please delete the directory manually: `/.paddlex/official_models/RT-DETR-L_wired_table_cell_det`.
Inference backend: onnxruntime
Inference backend config: cpu_num_threads=10
[INFO] ultra_infer/runtime/runtime.cc(308)::CreateOrtBackend Runtime initialized with Backend::ORT in Device::CPU.
Creating model: ('RT-DETR-L_wireless_table_cell_det', None)
Model files already exist. Using cached files. To redownload, please delete the directory manually: `/.paddlex/official_models/RT-DETR-L_wireless_table_cell_det`.
Inference backend: onnxruntime
Inference backend config: cpu_num_threads=10
[INFO] ultra_infer/runtime/runtime.cc(308)::CreateOrtBackend Runtime initialized with Backend::ORT in Device::CPU.
Creating model: ('PP-FormulaNet_plus-L', None)
Model files already exist. Using cached files. To redownload, please delete the directory manually: `/.paddlex/official_models/PP-FormulaNet_plus-L`.
Inference backend: onnxruntime
Inference backend config: cpu_num_threads=10
[INFO] ultra_infer/runtime/runtime.cc(308)::CreateOrtBackend Runtime initialized with Backend::ORT in Device::CPU.
Creating model: ('PP-Chart2Table', None)
Model files already exist. Using cached files. To redownload, please delete the directory manually: `/.paddlex/official_models/PP-Chart2Table`.
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
/usr/local/lib/python3.10/dist-packages/paddlex/inference/models/doc_vlm/predictor.py:100: UserWarning: The PP-Chart2Table series does not support `use_hpip=True` for now.
warnings.warn(
Loading configuration file /.paddlex/official_models/PP-Chart2Table/config.json
Loading weights file /.paddlex/official_models/PP-Chart2Table/model_state.pdparams
Loaded weights file from disk, setting weights to model.
All model checkpoint weights were used when initializing PPChart2TableInference.
All the weights of PPChart2TableInference were initialized from the model checkpoint at /.paddlex/official_models/PP-Chart2Table.
If your task is similar to the task the model of the checkpoint was trained on, you can already use PPChart2TableInference for predictions without further training.
Loading configuration file /.paddlex/official_models/PP-Chart2Table/generation_config.json
Traceback (most recent call last):
File "/usr/local/bin/paddlex", line 8, in <module>
sys.exit(console_entry())
File "/usr/local/lib/python3.10/dist-packages/paddlex/__main__.py", line 26, in console_entry
main()
File "/usr/local/lib/python3.10/dist-packages/paddlex/paddlex_cli.py", line 509, in main
pipeline_predict(
File "/usr/local/lib/python3.10/dist-packages/paddlex/paddlex_cli.py", line 370, in pipeline_predict
for res in result:
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/pipelines/_parallel.py", line 129, in predict
yield from self._pipeline.predict(
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/pipelines/layout_parsing/pipeline_v2.py", line 1007, in predict
layout_det_results = list(
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/models/base/predictor/base_predictor.py", line 219, in __call__
yield from self.apply(input, **kwargs)
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/models/base/predictor/base_predictor.py", line 277, in apply
prediction = self.process(batch_data, **kwargs)
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/models/object_detection/predictor.py", line 234, in process
batch_preds = self.infer(batch_inputs)
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/models/common/static_infer.py", line 641, in __call__
return self._call_multi_backend_infer(x)
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/models/common/static_infer.py", line 654, in _call_multi_backend_infer
return self._multi_backend_infer(inputs)
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/models/common/static_infer.py", line 594, in __call__
outputs = self.ui_runtime.infer(x)
File "/usr/local/lib/python3.10/dist-packages/ultra_infer/runtime.py", line 65, in infer
return self._runtime.infer(data)
RuntimeError: Exception from src/inference/src/cpp/infer_request.cpp:79:
Exception from src/inference/src/cpp/infer_request.cpp:66:
Exception from src/plugins/intel_cpu/src/infer_request.cpp:377:
Can't set the input tensor with index: 0, because the model input (shape=[1,3,640,640]) and the tensor (shape=(6.3.640.640)) are incompatible
当PP-DocLayout-M模型推理后端使用onnxruntime时：
（重复内容略）
Creating model: ('PP-DocLayout-M', None)
Model files already exist. Using cached files. To redownload, please delete the directory manually: `/.paddlex/official_models/PP-DocLayout-M`.
Inference backend: onnxruntime
Inference backend config: cpu_num_threads=10
[INFO] ultra_infer/runtime/runtime.cc(308)::CreateOrtBackend Runtime initialized with Backend::ORT in Device::CPU.
（重复内容略）
All the weights of PPChart2TableInference were initialized from the model checkpoint at /.paddlex/official_models/PP-Chart2Table.
If your task is similar to the task the model of the checkpoint was trained on, you can already use PPChart2TableInference for predictions without further training.
Loading configuration file /.paddlex/official_models/PP-Chart2Table/generation_config.json
[ERROR] ultra_infer/runtime/backends/ort/ort_backend.cc(378)::Infer Failed to Infer: Got invalid dimensions for input: image for the following indices
index: 0 Got: 6 Expected: 1
Please fix either the inputs/outputs or the model.
Traceback (most recent call last):
File "/usr/local/bin/paddlex", line 8, in <module>
sys.exit(console_entry())
File "/usr/local/lib/python3.10/dist-packages/paddlex/__main__.py", line 26, in console_entry
main()
File "/usr/local/lib/python3.10/dist-packages/paddlex/paddlex_cli.py", line 509, in main
pipeline_predict(
File "/usr/local/lib/python3.10/dist-packages/paddlex/paddlex_cli.py", line 370, in pipeline_predict
for res in result:
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/pipelines/_parallel.py", line 129, in predict
yield from self._pipeline.predict(
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/pipelines/layout_parsing/pipeline_v2.py", line 1007, in predict
layout_det_results = list(
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/models/base/predictor/base_predictor.py", line 219, in __call__
yield from self.apply(input, **kwargs)
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/models/base/predictor/base_predictor.py", line 278, in apply
prediction = PredictionWrap(prediction, len(batch_data))
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/models/base/predictor/base_predictor.py", line 56, in __init__
assert len(data[k]) == num, f"{len(data[k])} != {num} for key {k}!"
AssertionError: 0 != 6 for key boxes!
环境
- 请提供您使用的PaddlePaddle、PaddleX版本号、Python版本号
PaddlePaddle为3.0.0，PaddleX为3.2.1，Python为3.10.13。具体信息如下：
python --version
Python 3.10.13
pip list
Package Version
--------------------- --------------------
aiohappyeyeballs 2.6.1
aiohttp 3.13.0
aiosignal 1.4.0
aistudio-sdk 0.3.8
annotated-types 0.7.0
anyio 4.1.0
astor 0.8.1
async-timeout 5.0.1
attrs 25.4.0
bce-python-sdk 0.9.46
cachetools 6.2.1
certifi 2019.11.28
chardet 3.0.4
charset-normalizer 3.4.4
click 8.3.0
coloredlogs 15.0.1
colorlog 6.9.0
cssselect 1.3.0
cssutils 2.11.1
dbus-python 1.2.16
decorator 5.1.1
distro-info 0.23+ubuntu1.1
einops 0.8.1
et_xmlfile 2.0.0
exceptiongroup 1.2.0
fastapi 0.119.0
filelock 3.20.0
filetype 1.2.0
flatbuffers 25.9.23
frozenlist 1.8.0
fsspec 2025.9.0
ftfy 6.3.1
future 1.0.0
h11 0.14.0
hf-xet 1.1.10
httpcore 1.0.2
httpx 0.25.1
huggingface-hub 0.35.3
humanfriendly 10.0
idna 2.8
imagesize 1.4.1
Jinja2 3.1.6
joblib 1.5.2
lxml 6.0.2
MarkupSafe 3.0.3
ml_dtypes 0.5.3
modelscope 1.30.0
more-itertools 10.8.0
mpmath 1.3.0
multidict 6.7.0
networkx 3.4.2
numpy 1.26.2
onnx 1.17.0
onnx_graphsurgeon 0.5.8
onnxoptimizer 0.3.13
onnxruntime 1.22.0
opencv-contrib-python 4.10.0.84
openpyxl 3.1.5
opt-einsum 3.3.0
packaging 25.0
paddle2onnx 2.0.2rc3
paddlepaddle 3.0.0
paddlex 3.2.1
pandas 2.3.3
Pillow 10.1.0
pip 23.3.1
polygraphy 0.49.26
premailer 3.10.0
prettytable 3.16.0
propcache 0.4.1
protobuf 4.25.1
psutil 7.1.0
py-cpuinfo 9.0.0
pyclipper 1.3.0.post6
pycryptodome 3.23.0
pydantic 2.12.1
pydantic_core 2.41.3
PyGObject 3.36.0
pypdfium2 4.30.0
python-apt 2.0.1+ubuntu0.20.4.1
python-dateutil 2.9.0.post0
pytz 2025.2
PyYAML 6.0.2
regex 2025.9.18
requests 2.32.5
requests-unixsocket 0.2.0
ruamel.yaml 0.18.15
ruamel.yaml.clib 0.2.14
scikit-learn 1.7.2
scipy 1.15.3
setuptools 68.2.2
shapely 2.1.2
six 1.14.0
sniffio 1.3.0
starlette 0.48.0
sympy 1.14.0
threadpoolctl 3.6.0
tiktoken 0.12.0
tokenizers 0.22.1
tqdm 4.67.1
typing_extensions 4.15.0
typing-inspection 0.4.2
tzdata 2025.2
ujson 5.11.0
ultra-infer-python 1.2.0
unattended-upgrades 0.1
urllib3 2.5.0
uvicorn 0.37.0
wcwidth 0.2.14
wheel 0.45.1
yarl 1.22.0
- 请提供您使用的操作系统信息，如Linux/Windows/MacOS
Linux任意发行版，目前在Debian 13，Ubuntu 22.04稳定复现。
- 请问您使用的CUDA/cuDNN的版本号是？
N/A（CPU推理模式）
Metadata
Metadata
Assignees
Labels
No labels