memect-ppx 0.0.1.post1

memect-ppx 是一个 PDF / 图片文档解析工具，将输入文件转换为结构化的 Markdown、JSON。支持本地模型（默认）和多种 LLM 后端（DeepSeek、PaddleOCR、GLM等），适用于高精度文档理解场景。

 PPX — High-Accuracy PDF & Image Parser

简体中文 | English

---

Convert PDF and images to structured Markdown / JSON — locally, accurately, production-ready.

PPX is an open-source document parsing engine built for high-fidelity extraction of text, tables, figures, formulas, and layout from PDFs and images. It ships with a built-in OCR + layout pipeline and optionally offloads recognition to state-of-the-art LLM backends (DeepSeek-OCR, PaddleOCR-VL, GLM-OCR).

• What output do I get? — Markdown and JSON; every object carries page coordinates.
• Do I need a GPU? — No. The default backend runs on CPU. GPU (CUDA) is optional for throughput.
• Does it handle scanned PDFs? — Yes. OCR is applied automatically when native text is absent.
• Can I use my own LLM? — Yes. Any OpenAI-compatible endpoint is accepted via --backend.
• Is it embeddable? — Yes. LGPL-3.0 lets you link it into commercial products.

---

Get Started in 30 Seconds

uv pip install memect-ppx onnxruntime opencv-contrib-python
# or
pip install memect-ppx onnxruntime opencv-contrib-python
ppx parse document.pdf -o output/

PPX uses the pipeline mode by default. The parsed Markdown is typically written to output/doc.md when -o output/ is provided.

---

What Problems Does This Solve?

Problem	How PPX Handles It
Native-text PDF with invisible/garbled characters	Detects encoding anomalies; falls back to OCR per page
Scanned document with no embedded text	Full-page OCR or vLLM backend
Complex table spanning multiple columns/rows	LLM-based structural parsing, colspan/rowspan preserved
Math-heavy academic paper	LaTeX formula extraction
Batch processing thousands of files	Directory-level parse dir/ with -o output/

---

Example Outputs

Mixed table content

This example shows a mixed table scenario where the table body contains editable text, while much of the header area is still image-based.

Input snippet:

Markdown output:

JSON output:

Scanned English table

This example shows a scanned English table parsing result.

Markdown output:

JSON output:

---

Benchmarks

See docs/BENCHMARKS.md for benchmark results, citation, attribution, and compliance notes.

---

Capability Matrix

Capability	Default (Local)	DeepSeek-OCR	PaddleOCR-VL	GLM-OCR
Text extraction	✅	✅	✅	✅
Per-character coordinates	✅	❌	❌	❌
Table structure (colspan / rowspan)	✅	✅	✅	✅
Formula → LaTeX	✅	✅	✅	✅
Figure region extraction	✅	✅	✅	✅
CPU-only mode	✅	✅	✅	✅
CUDA acceleration	✅	✅	✅	✅
No external service required	✅	❌	❌	❌

---

Which Backend Should I Use?

Scenario	Recommended Backend
Privacy-sensitive documents, air-gapped environment	default
Highest accuracy on complex layouts	deepseek
Good accuracy, lighter GPU footprint (~10 GB)	paddle
Fast inference with speculative decoding	glm
Quick integration test / CI pipeline	default (CPU)

---

Quick Start

Default pipeline mode

ppx parse <input_path> -o <output_path>

# Example
ppx parse report.pdf -o output/

Parse a single file

# Auto-detect whether OCR is needed
ppx parse report.pdf

# Force OCR on every page
ppx parse report.pdf --ocr yes

# Skip OCR entirely
ppx parse report.pdf --ocr no

# Parse an image
ppx parse scan.png

Batch processing

# Parse all PDFs and images in a directory
ppx parse docs/

# Write output to a specific directory
ppx parse docs/ -o output/

Use an LLM backend

# DeepSeek-OCR (requires ~20 GB VRAM via vLLM)
ppx parse report.pdf --backend deepseek \
  --deepseek '{"base_url":"http://127.0.0.1:4000/v1","model":"deepseek-ocr-2","api_key":""}'

# PaddleOCR-VL (requires ~10 GB VRAM)
ppx parse report.pdf --backend paddle \
  --paddle '{"base_url":"http://127.0.0.1:4001/v1","model":"paddleocr-vl","api_key":""}'

# GLM-OCR (requires ~10 GB VRAM)
ppx parse report.pdf --backend glm \
  --glm '{"base_url":"http://127.0.0.1:4002/v1","model":"glmocr","api_key":""}'

Persist configuration

Tired of typing the same flags? Drop a config file:

mkdir conf
# conf/settings.py  (Python dict) or conf/settings.json
# Reference: src/memect/conf/settings.custom.py

# conf/settings.py
settings = {
    "pdf_parser.deepseek.model.base_url": "http://127.0.0.1:4000/v1",
    "pdf_parser.paddle.model.base_url": "http://127.0.0.1:4001/v1",
    "pdf_parser.glm.model.base_url": "http://127.0.0.1:4002/v1",
}

Now just run:

ppx parse report.pdf --backend deepseek

---

Use from python

PPX can be used directly as a library. If you call it repeatedly, a single global Parser instance is usually enough.

from memect.pdf.parser import Parser
from memect.pdf.base import KDocument, KDocumentFactory

# If you call it repeatedly, a single global parser is usually enough.
# If no arguments are passed, the default settings are used.
with Parser() as parser:
    doc = KDocument("/path/your.pdf")
    parser.parse(doc)

# Batch parsing with multiprocessing and default settings.
doc = KDocumentFactory("/path/your.pdf", params=None)
docs = [doc]
Parser.batch(docs, max_workers=1)

---

CLI Reference

ppx parse <path> [OPTIONS]

Arguments:
  path          PDF file, image file, or directory

Options:
  --backend     default | deepseek | paddle | glm   (default: default)
  --ocr         yes | no | auto                      (default: auto)
  --table       no | ybk | wbk | auto | llm          (default: auto)
  --pages       Page range, e.g. "1-5,10"
  --mode        page | tree                    (default: page)
  -o, --output  Output directory

Other subcommands:

ppx start               Launch HTTP API server

---

Output Format

Each parsed document is written to <input>.out/:

report.pdf.out/
├── doc.md          # full document in Markdown
├── doc.json        # full structured data with per-object coordinates
├── pages/          # per-page breakdown (one entry per page)
└── images/         # extracted figures/images (present when figures are detected)

Path	Description
doc.md	Markdown with figure references
doc.json	JSON tree: document → pages → objects, each with bounding-box coordinates
pages/	Per-page Markdown and JSON, useful for page-level processing
images/	Extracted image regions; only present when the document contains figures

---

Installation

Option A — PyPI (recommended)

# Create a virtual environment
uv venv -p 3.12
source .venv/bin/activate

# CPU build
uv pip install memect-ppx
uv pip install onnxruntime --no-config
uv pip install opencv-contrib-python --no-config   # or opencv-contrib-python-headless

# GPU (CUDA) build
uv pip install memect-ppx[cuda]
uv pip install onnxruntime-gpu --no-config
uv pip install opencv-contrib-python --no-config

Why install onnxruntime and opencv manually? Third-party packages often pin different variants (headless vs contrib, cpu vs gpu). PPX excludes both from its dependency list so you stay in control of which variant is installed.

Option B — From source

git clone https://github.com/memect/memect-ppx.git
cd ppx
uv venv -p 3.12
source .venv/bin/activate

# Install all dependencies (CPU)
uv sync --no-install-project
uv pip install onnxruntime --no-config
uv pip install opencv-contrib-python --no-config

# Or GPU
uv sync --extra cuda --no-install-project
uv pip install onnxruntime-gpu --no-config
uv pip install opencv-contrib-python --no-config

---

Platform Support

Platform	Python	CPU	CUDA	Notes
Linux	>= 3.12	✅	✅	Recommended for production
macOS (Apple Silicon)	>= 3.12	✅	❌
macOS (Intel)	3.12 – 3.13	✅	❌	Capped by OpenVINO
Windows	>= 3.12	✅	✅	Community-tested

CUDA requires NVIDIA driver + CUDA 12.x and onnxruntime-gpu built for that CUDA version.

---

Launching LLM Services

PPX LLM backends are served via vLLM.

DeepSeek-OCR-2 (~20 GB VRAM)

vllm serve ./hub/deepseek-ai/DeepSeek-OCR-2 \
  --served-model-name deepseek-ocr-2 \
  --logits-processors vllm.model_executor.models.deepseek_ocr:NGramPerReqLogitsProcessor \
  --mm-processor-cache-gb 0 \
  --no-enable-prefix-caching \
  --gpu-memory-utilization 0.8 \
  --port 4000

PaddleOCR-VL / PaddleOCR-VL-1.5 (~10 GB VRAM)

vllm serve ./hub/PaddlePaddle/PaddleOCR-VL \
  --served-model-name paddleocr-vl \
  --trust-remote-code \
  --max-num-batched-tokens 16384 \
  --no-enable-prefix-caching \
  --mm-processor-cache-gb 0 \
  --gpu-memory-utilization 0.5 \
  --port 4001

Replace PaddleOCR-VL with PaddleOCR-VL-1.5 to use the newer model; port and --served-model-name remain the same.

GLM-OCR (~10 GB VRAM)

# Requires transformers >= 5.3.0
uv pip install "transformers>=5.3.0"

vllm serve ./hub/ZhipuAI/GLM-OCR \
  --served-model-name glmocr \
  --max-num-batched-tokens 16384 \
  --max-model-len 16384 \
  --speculative-config '{"method": "mtp", "num_speculative_tokens": 1}' \
  --gpu-memory-utilization 0.5 \
  --port 4002

Model source: ModelScope — ZhipuAI/GLM-OCR

FAQ

Does PPX support password-protected PDFs?

Not currently. Strip the password with a tool like qpdf before passing the file to PPX.

How do I resolve opencv version conflicts?

Uninstall all existing opencv variants first, then reinstall:

uv pip uninstall opencv-python opencv-contrib-python \
                  opencv-python-headless opencv-contrib-python-headless
uv pip install opencv-contrib-python --no-config

Can onnxruntime and onnxruntime-gpu coexist?

No. Install exactly one. The GPU variant must match your system's CUDA version.

Can I use PPX on Mac with GPU acceleration?

No. Neither Apple Silicon nor Intel Macs support CUDA. The CPU backend works on both.

Can I embed PPX in a commercial product?

Yes. LGPL-3.0 permits linking PPX as a library in proprietary software. Modifications to PPX's own source must be released under LGPL-3.0.

How do I parse only specific pages?

ppx parse report.pdf --pages "1-5,10,15-20"

---

Contributing

We welcome bug reports, feature requests, and pull requests.

1. Fork the repository and create a feature branch.
2. Run tests: uv run pytest
3. Submit a PR — please describe the motivation and include test cases.

See CONTRIBUTING.md for full guidelines.

---

License

PPX is released under the GNU Lesser General Public License v3.0 (LGPL-3.0).

LGPL-3.0 allows you to link this library into your applications — including commercial ones — without relicensing your own code. Modifications to PPX itself must be shared under the same license.

For bundled third-party code and assets, see NOTICE and docs/THIRD_PARTY_LICENSES.md. Those files document attribution and redistribution review items for vendored components and bundled resources shipped with this repository.

Product site: https://pdf2x.cn/

Mini Program code: