YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
MagiV3 ONNX Mobile Export
This folder contains community-converted ONNX artifacts for mobile/runtime experimentation.
Credit and Attribution
This conversion is based on the original Magi research and model development by Ragav Sachdeva and Andrew Zisserman (with co-authors where applicable).
All core model architecture, training, and research contributions belong to the original authors.
Original Hugging Face model/repository:
Important Note
These are converted/exported files only (ONNX + tokenizer + external data shards).
This is not an official release from the original authors.
For this export, you must keep all files together in the same model directory:
magiv3-forward.onnxtokenizer.json- all
model.*shard files
If any shard file is missing, ONNX Runtime loading will fail.
Citations
@InProceedings{magiv1,
author = {Sachdeva, Ragav and Zisserman, Andrew},
title = {The Manga Whisperer: Automatically Generating Transcriptions for Comics},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2024},
pages = {12967-12976}
}
@InProceedings{Sachdeva_2024_ACCV,
author = {Sachdeva, Ragav and Shin, Gyungin and Zisserman, Andrew},
title = {Tails Tell Tales: Chapter-wide Manga Transcriptions with Character Names},
booktitle = {Proceedings of the Asian Conference on Computer Vision (ACCV)},
month = {December},
year = {2024},
pages = {2053-2069}
}
@misc{sachdeva2025panelsprosegeneratingliterary,
title={From Panels to Prose: Generating Literary Narratives from Comics},
author={Ragav Sachdeva and Andrew Zisserman},
year={2025},
eprint={2503.23344},
archivePrefix={arXiv},
primaryClass={cs.CV},
url={https://arxiv.org/abs/2503.23344},
}
License
Please follow the original model/repository license and usage terms from: