zjunlp/ModelKinship

Exploring Model Kinship for Merging Large Language Models

35
/ 100
Emerging

This tool helps AI researchers and practitioners who merge large language models (LLMs) to understand how different models relate to each other. It takes two fine-tuned LLMs and their common base model, then calculates a 'kinship' score. This score indicates the similarity in how the models learned, helping users choose optimal models for merging to achieve better performance and reduce trial-and-error.

No commits in the last 6 months.

Use this if you are iteratively merging large language models and want to make more informed decisions about which models to combine for optimal results.

Not ideal if you are looking for a general-purpose LLM evaluation or merging toolkit, as this focuses specifically on analyzing model relatedness.

large-language-models model-merging AI-research LLM-optimization
Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 7 / 25
Maturity 16 / 25
Community 10 / 25

How are scores calculated?

Stars

27

Forks

3

Language

Python

License

MIT

Last pushed

Apr 16, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/transformers/zjunlp/ModelKinship"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.