Tools,

PP-OCRv5 mobile vs server, tested

1,904

of 1,955 words: mobile, English model, ONNX

Get the code

Server lost to mobile on book pages

Two scanned English book pages, a Mac CPU, 2 threads a run.

Words found of 381, and time a page
Models Words Time
Mobile, English model, PaddleOCR 360 41 s
Mobile, English model, RapidOCR 358 8 s
Mobile, RapidOCR 350 9 s
Mobile, PaddleOCR 349 45 s
Server, PaddleOCR, 960 px cap 345 39 s
Server, RapidOCR 341 104 s
Server, PaddleOCR 337 129 s

PaddleOCR's own recognition table agrees on English: mobile 66.00, server 64.70.

PaddleOCR server took 36 GB by default

PaddleOCR's default is type min with limit_side_len 64. Type min only scales small pages up. So detection got our 1920 px pages at full size.

There the server models peaked at 36 GB of memory and took 129 s a page. A cap on the long side fixed most of it:

text_det_limit_type="max", text_det_limit_side_len=960

With the cap: 7.3 GB, 39 s a page, and 8 more words.

The English model read 1,904 of 1,955

All 10 pages, mobile only. PaddleOCR server would have taken 21 minutes.

Words found of 1,955, and time a page
Setup Words Time
v5 mobile, English model, RapidOCR 1,904 8 s
PP-OCRv6 small, RapidOCR default 1,895 4.5 s
Our Actor's code 1,879 3.5 s
Tesseract 5.5.4 1,878 3.1 s
v5 mobile, RapidOCR 1,869 8 s
v5 mobile, PaddleOCR 1,856 47 s

Screenshots: the English model won again

Words right on 16 screenshots, 1,938 words
Setup Words right
v5 mobile, English model, RapidOCR 97.5%
Apple Vision 96.9%
v5 mobile, PaddleOCR 94.9%
Our Actor's code 94.4%
v5 mobile, RapidOCR 93.2%
v5 server, RapidOCR 76.0%

In RapidOCR, the server models glued English words, like youhavetobein theoficeorat. PaddleOCR's server run read that line right.

RapidOCR ran mobile 6 times faster

Pick mobile by name in PaddleOCR

from paddleocr import PaddleOCR

ocr = PaddleOCR(
    text_detection_model_name="PP-OCRv5_mobile_det",
    text_recognition_model_name="en_PP-OCRv5_mobile_rec",
    use_doc_orientation_classify=False,
    use_doc_unwarping=False,
    use_textline_orientation=False,
)
for res in ocr.predict("page.jpg"):
    print("\n".join(res["rec_texts"]))

lang="en", ocr_version="PP-OCRv5" gives you the server detector with the English model. No language setting picks the mobile detector.

Or run the ONNX files in RapidOCR

pip install rapidocr onnxruntime
from rapidocr import LangRec, ModelType, OCRVersion, RapidOCR

engine = RapidOCR(
    params={
        "Det.ocr_version": OCRVersion.PPOCRV5,
        "Det.model_type": ModelType.MOBILE,
        "Rec.ocr_version": OCRVersion.PPOCRV5,
        "Rec.model_type": ModelType.MOBILE,
        "Rec.lang_type": LangRec.EN,
    }
)
print("\n".join(engine("page.jpg").txts))

On our first page, both scripts gave the same text as the runs above. RapidOCR has no English server model.

Five models and their sizes

Download size, PaddleOCR folder and RapidOCR ONNX file
Model Paddle ONNX
PP-OCRv5_mobile_det 4.9 MB 4.8 MB
PP-OCRv5_server_det 88 MB 88 MB
PP-OCRv5_mobile_rec 17 MB 17 MB
PP-OCRv5_server_rec 85 MB 85 MB
en_PP-OCRv5_mobile_rec 8.0 MB 7.9 MB

All five are Apache-2.0. The two main recognizers read Chinese, Traditional Chinese, English and Japanese. The English one reads English only.

How we ran it
  • A Mac with Apple M5 Max, macOS 26.6. PaddleOCR 3.7.0 with paddlepaddle 3.3.1, and RapidOCR 3.10.0 with onnxruntime 1.32.0. Most runs: two at a time, 2 threads each, at low priority.
  • The scan: 10 proofread pages from two public domain books on Wikisource, 1920 px JPEGs, 1,955 words. The 381-word table uses its first 2 pages.
  • PaddleOCR ran with its three page-fix models off. RapidOCR ran with its defaults, which include its line-turn model and a 2,000 px cap on the long side, plus the v5 model names.
  • Score: true words found in order. Screenshots: the 16 at 1x from our screenshot test, words right in any order.
  • Times are rough: the Mac ran other work too. The Tesseract, Actor and Apple Vision rows are from our earlier runs today.
  • I stopped the PaddleOCR server run on screenshots after 6 of 16 shots, to keep the daytime test short. RapidOCR server with the 960 px cap found 342 words in 109 s a page.

Scanned pages to text, no install

Run it yourself. Our Apify Actor turns a scanned PDF into Markdown or text. $0.006 a scanned page. Its code found 1,879 of these 1,955 words.

Try the Actor

Or send us the scan. Tack, our AI, does the job for $29, up to 50 pages. First page free.

Get one page free

Questions people ask

Is PP-OCRv5 server better than mobile?

Not on our English pages. Mobile found more words on book scans and on screenshots, and it was faster. PaddleOCR's own tables favor server for Chinese and handwriting.

Which PP-OCRv5 model is best for English?

Of the pairs we tested, en_PP-OCRv5_mobile_rec with PP-OCRv5_mobile_det. In RapidOCR it found 1,904 of 1,955 words on our scan and 97.5% of the words on 16 screenshots.

Where do I get PP-OCRv5 ONNX files?

RapidOCR downloads them for you from ModelScope on first use. PaddlePaddle also posts ONNX copies on Hugging Face, such as PaddlePaddle/PP-OCRv5_mobile_det_onnx. Both are Apache-2.0.

Is PP-OCRv5 still the default?

No. PaddleOCR 3.7.0 uses PP-OCRv6 medium by default, and RapidOCR 3.10.0 uses PP-OCRv6 small. Both still run PP-OCRv5 when you name it.

Does your Actor use PP-OCRv5?

No. It reads scans with Tesseract. Its second reader, for English pages that Tesseract reads badly, is RapidOCR with PP-OCRv6 small.