5 1

Jason Ni

jasonni2

AI & ML interests

None yet

Recent Activity

new activity 1 day ago

alex-dinh/PP-DocLayoutV3-ONNX:Order seems to be by score

new activity about 1 month ago

Mungert/MinerU2.5-2509-1.2B-GGUF:llama.cpp example

new activity about 2 months ago

tencent/HY-MT1.5-1.8B-GGUF:When using ollama to run this model, the output is "onse", you can use the TEMPLATE below to correct it

View all activity

Organizations

New activity in alex-dinh/PP-DocLayoutV3-ONNX 1 day ago

Order seems to be by score

#1 opened 3 days ago by

gstrat88

New activity in Mungert/MinerU2.5-2509-1.2B-GGUF about 1 month ago

llama.cpp example

#1 opened 4 months ago by

engrtipusultan

New activity in tencent/HY-MT1.5-1.8B-GGUF about 2 months ago

When using ollama to run this model, the output is "onse", you can use the TEMPLATE below to correct it

👍 2

#6 opened about 2 months ago by

hejiangda

New activity in lightonai/LightOnOCR-1B-1025 3 months ago

Without bounding box output

#11 opened 3 months ago by

jasonni2

commented on SOTA OCR with Core ML and dots.ocr 4 months ago

Thank your for trying it. Rest API is one item on the TODO List. I'm working on parsing PDF file.

commented on SOTA OCR with Core ML and dots.ocr 4 months ago

For guys can't wait for it any more, please have a try of my Dots.OCR.Runner https://apps.apple.com/us/app/dots-ocr-runner/id6753667495

It's implemented based on llama.cpp. With my experience of this implementation, I think the most challenging part is the vision encoder that has 42 layers, which consumes quite amount of memory. With ggml framework, we can manually control the inference flow to limit memory peek.

This article is very helpful to me. Looking forward to the upcomming parts!

upvoted an article 5 months ago

Article

SOTA OCR with Core ML and dots.ocr

Oct 2, 2025

•

New activity in nvidia/parakeet-tdt-0.6b-v2 7 months ago

ONNX conversion

#9 opened 10 months ago by

Berrisius