rag-chunker models
ONNX models that rag-chunker downloads on first use for local page layout analysis, OCR and table structure recognition. They are unmodified exports of PaddleOCR / PP-StructureV3 models (Apache-2.0); this repository only keeps them at a stable address with their checksums, so the tool keeps working whatever happens to the original mirrors.
rag-chunker verifies the sha256 of every download against the catalogue in its source
(src/local/models.ts), so a changed file is rejected.
| File | What it is | Size | Licence | Original | sha256 |
|---|---|---|---|---|---|
PP-DocLayoutV3.onnx |
Page layout: 25 block classes with reading order | 123.9 MB | Apache-2.0 (PaddlePaddle PP-DocLayoutV3) | source | dc5670ebbb42e2ba4e41395fc55e8217b007134e8b9e35023d592a8fe040a288 |
PP-OCRv5_mobile_det_infer.onnx |
OCR text detection (DBNet, mobile) | 4.5 MB | Apache-2.0 (PaddleOCR PP-OCRv5) | source | d7fe3ea74652890722c0f4d02458b7261d9f5ae6c92904d05707c9eb155c7924 |
PP-OCRv5_server_det_infer.onnx |
OCR text detection (DBNet, server: larger backbone, finds more small text) | 83.7 MB | Apache-2.0 (PaddleOCR PP-OCRv5) | source | 7c1a843c497972b9b470748e91a6d83c0f79c4b5b6459508ac533a88f418f33b |
latin_PP-OCRv5_mobile_rec_infer.onnx |
OCR text recognition, Latin scripts (Turkish, English, German, French, Spanish, …) | 7.7 MB | Apache-2.0 (PaddleOCR PP-OCRv5) | source | 497dbed20b7fd86334c9deb5082c2958982a316bb16e3589ebc6502cd85cae79 |
ppocrv5_latin_dict.txt |
Character table of the Latin recognition model | 0.0 MB | Apache-2.0 (PaddleOCR PP-OCRv5) | source | 7274e68c7675355e45dd75c360c83faa0d0624de33a704c799a7da8897662201 |
PP-OCRv5_server_rec_infer.onnx |
OCR text recognition, server: Chinese, Japanese, English and most Latin accents (no Ğ/İ) | 80.3 MB | Apache-2.0 (PaddleOCR PP-OCRv5) | source | c8cff87fd83a76b97178250653924598593088c6bb71b2d2d942ef13b9a2d2ec |
ppocrv5_dict.txt |
Character table of the server recognition model (18,383 characters) | 0.1 MB | Apache-2.0 (PaddleOCR PP-OCRv5) | source | d1979e9f794c464c0d2e0b70a7fe14dd978e9dc644c0e71f14158cdf8342af1b |
SLANet_plus.onnx |
Table structure: rows, columns, merged cells and cell boxes from a table image | 7.4 MB | Apache-2.0 (PaddleOCR PP-StructureV2) | source | d57a942af6a2f57d6a4a0372573c696a2379bf5857c45e2ac69993f3b334514b |
table_structure_dict_ch.txt |
Structure tokens of the SLANet models | 0.0 MB | Apache-2.0 (PaddleOCR) | source | 68d344a84b726e043f390122240ff2b2ced2949b2a80ce9b61ae955054d190ef |
PP-LCNet_x1_0_doc_ori.onnx |
Page orientation: upright, 90, 180 or 270 degrees | 6.5 MB | Apache-2.0 (PaddleOCR PP-LCNet_x1_0_doc_ori) | source | 6039f8048d654587addc5e8c94cf89573a1b07af63e92b20691de98ab68deeb5 |
texify2/config.json |
Formula recognition (texify2): model configuration | 0.0 MB | CC-BY-SA-4.0 (vikp/texify2) | source | 7133e06cbdbb942886cd5595b860596d08a8dd49c0e91ae7e14283f18fe8b127 |
texify2/generation_config.json |
Formula recognition (texify2): generation settings | 0.0 MB | CC-BY-SA-4.0 (vikp/texify2) | source | 0fb8980236d664d2851ed5939b435ecc6fe0e32ee08b83e75cb1743db0cfe10e |
texify2/preprocessor_config.json |
Formula recognition (texify2): image preprocessing settings | 0.0 MB | CC-BY-SA-4.0 (vikp/texify2) | source | b2fc6ff0098ac8e6af6e688bfc28188db883c99a48afa60e8555faeedc0c45e4 |
texify2/tokenizer.json |
Formula recognition (texify2): tokenizer | 2.0 MB | CC-BY-SA-4.0 (vikp/texify2) | source | 02c318d9cfa95bf323371762b8f838a82709530274d36dba6eca880f0add6cc4 |
texify2/tokenizer_config.json |
Formula recognition (texify2): tokenizer settings | 0.0 MB | CC-BY-SA-4.0 (vikp/texify2) | source | fd4d94f8b9dbb7deeb3a3ef084ca0e16c43d45774a180730b7e9cfd6359a074b |
texify2/onnx/encoder_model_quantized.onnx |
Formula recognition (texify2): image encoder, 8-bit | 75.6 MB | CC-BY-SA-4.0 (vikp/texify2) | source | 6ef545e9b9c451951077ee8f54282fcc77156380df7c3ce5c14852102c921397 |
texify2/onnx/decoder_model_merged_quantized.onnx |
Formula recognition (texify2): LaTeX decoder, 8-bit | 228.8 MB | CC-BY-SA-4.0 (vikp/texify2) | source | c9d9f2b65d8c0ea9ff1b3c3ba54a98c1d93f91c690ac10877a395bb7a8cd407a |
Usage
Nothing to do by hand: npx @msbayindir/rag-chunker process paper.pdf --layout --ocr pp-ocr
fetches what it needs into ~/.rag-chunker/models (RAG_CHUNKER_MODELS_DIR changes the folder).
To use the files directly, load them with onnxruntime; the pre- and post-processing rag-chunker
applies (input sizes, normalisation, dictionaries, decoding) is in src/local/layout.ts,
src/local/ocr.ts and src/local/table.ts.
Credits and licences
The models were trained and released by the PaddlePaddle / PaddleOCR team under the Apache License 2.0
(see LICENSE and NOTICE). The ONNX exports come from https://huggingface.co/msbayindir/rag-chunker-models/resolve/main and https://huggingface.co/msbayindir/rag-chunker-models/resolve/main/texify2 and https://huggingface.co/msbayindir/rag-chunker-models/resolve/main/texify2/onnx.
This mirror adds nothing of its own.