- Libraries
- LiteRT-LM
How to use litert-community/MiniCPM-OCR-1.0 with LiteRT-LM:
# LiteRT-LM runs on various platforms (Android, iOS, Windows, Linux, macOS, IoT, Web/WASM)
# and supports many APIs (C++, Python, Kotlin, Swift, JavaScript, Flutter).
# For platform-specific integration guides, please refer to the official developer website:
# https://ai.google.dev/edge/litert-lm
# To try LiteRT-LM, the easiest way is to use our CLI tool.
# 1. Install the LiteRT-LM CLI tool:
pip install -U litert-lm
# 2. Download and run this model locally:
# See: https://ai.google.dev/edge/litert-lm/cli
# A single .litertlm file in the repo is picked automatically; otherwise the CLI asks which one to run
# (or pass its name right after the repo id).
litert-lm run \
--from-huggingface-repo=litert-community/MiniCPM-OCR-1.0 \
--prompt="Write me a poem"
- LiteRT
How to use litert-community/MiniCPM-OCR-1.0 with LiteRT:
# No code snippets available yet for this library.
# To use this model, check the repository files and the library's documentation.
# Want to help? PRs adding snippets are welcome at:
# https://github.com/huggingface/huggingface.js
- MiniCPM-OCR
How to use litert-community/MiniCPM-OCR-1.0 with MiniCPM-OCR:
pip install "minicpm-ocr[all]" huggingface_hub
# Deploy litert-community/MiniCPM-OCR-1.0 as a recognition service before running the Python example:
# https://github.com/OpenBMB/MiniCPM-OCR/blob/main/docs/vllm_deploy_en.md
# vLLM: load the model above; use --port 8080,
# --served-model-name minicpm-ocr and --max-model-len 40960.
# Layout detection uses cuda:0; recognition runs on the deployed service.
from importlib.resources import files
from pathlib import Path
from minicpmvocr import load_config, pipelined_run
from huggingface_hub import snapshot_download
layout_model = snapshot_download("PaddlePaddle/PP-DocLayoutV3")
cfg = load_config(
str(files("minicpmvocr").joinpath("configs/minicpmvocr.yaml")),
overrides={
"input": {"file_path": str(Path("demo.png").resolve())},
"engine": {
"type": "vllm",
"server_url": "http://127.0.0.1:8080",
"model_name": "minicpm-ocr",
"max_model_len": 40960,
},
"layout": {"model_dir": layout_model, "device": "cuda:0"},
# Render PDFs in-process so this example also works without a main guard.
"pipeline": {"overlap_layout": True, "render_workers": 1},
},
)
pipelined_run(cfg) - Notebooks
- Google Colab
- Kaggle
