zjunlp/OneGen

[EMNLP 2024] OneGen: Efficient One-Pass Unified Generation and Retrieval for LLMs.

39
/ 100
Emerging

OneGen helps AI developers efficiently fine-tune large language models (LLMs) for tasks that involve both generating text and finding relevant information. It takes an LLM and training data for specific tasks like answering questions or linking entities, and produces a single, optimized LLM ready for deployment. This is for AI/ML engineers and researchers who are building applications that require LLMs to generate accurate responses by retrieving information.

147 stars. No commits in the last 6 months.

Use this if you need to fine-tune LLMs for retrieval-augmented generation (RAG) tasks and want to reduce model deployment and inference costs by using a single model.

Not ideal if you are looking for an out-of-the-box, no-code solution for end-user applications, as this project requires developer expertise to implement.

LLM fine-tuning Generative AI development Information retrieval Natural Language Processing AI infrastructure optimization
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 13 / 25

How are scores calculated?

Stars

147

Forks

15

Language

Python

License

MIT

Last pushed

Nov 13, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/rag/zjunlp/OneGen"

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