modelscope/easydistill
a toolkit on knowledge distillation for large language models
This project helps AI researchers and industry practitioners make large language models (LLMs) more efficient. It takes an existing, powerful LLM and a smaller, target LLM, then trains the smaller model to mimic the performance of the larger one using various techniques. The output is a smaller, faster LLM that performs nearly as well as its much larger counterpart, ideal for deployment where computational resources are limited.
292 stars.
Use this if you need to deploy powerful large language models in environments with limited computing resources, or if you want to reduce the cost and latency of using LLMs without sacrificing accuracy.
Not ideal if you are looking to train a large language model from scratch, or if your primary goal is to develop new LLM architectures rather than optimize existing ones.
Stars
292
Forks
31
Language
Python
License
Apache-2.0
Category
Last pushed
Mar 10, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/modelscope/easydistill"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Compare
Related models
scaleapi/llm-engine
Scale LLM Engine public repository
AGI-Arena/MARS
The official implementation of MARS: Unleashing the Power of Variance Reduction for Training Large Models
AGI-Edgerunners/LLM-Adapters
Code for our EMNLP 2023 Paper: "LLM-Adapters: An Adapter Family for Parameter-Efficient...
Wang-ML-Lab/bayesian-peft
Bayesian Low-Rank Adaptation of LLMs: BLoB [NeurIPS 2024] and TFB [NeurIPS 2025]
sangmichaelxie/doremi
Pytorch implementation of DoReMi, a method for optimizing the data mixture weights in language...