eliahuhorwitz/MoTHer
Official PyTorch Implementation for the "Unsupervised Model Tree Heritage Recovery" paper (ICLR 2025).
This project helps machine learning practitioners and researchers understand the lineage of AI models. You input a collection of neural network models, and it outputs a 'Model Tree' or 'Model Graph' showing which models were fine-tuned from others and in what order. This is useful for anyone who needs to trace the heritage of publicly shared AI models, especially for intellectual property or transparency purposes.
No commits in the last 6 months.
Use this if you need to automatically discover the historical relationships between a set of neural network models, for example, to understand their origins or verify compliance.
Not ideal if you already have explicit documentation of model heritage or if you are working with non-neural network model types.
Stars
63
Forks
—
Language
Python
License
—
Category
Last pushed
Jul 01, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/eliahuhorwitz/MoTHer"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
ModelCloud/GPTQModel
LLM model quantization (compression) toolkit with hw acceleration support for Nvidia CUDA, AMD...
intel/auto-round
🎯An accuracy-first, highly efficient quantization toolkit for LLMs, designed to minimize quality...
pytorch/ao
PyTorch native quantization and sparsity for training and inference
bodaay/HuggingFaceModelDownloader
Simple go utility to download HuggingFace Models and Datasets
NVIDIA/kvpress
LLM KV cache compression made easy