aniquetahir/JORA

JORA: JAX Tensor-Parallel LoRA Library (ACL 2024)

26
/ 100
Experimental

This tool helps machine learning engineers and researchers efficiently fine-tune large language models, specifically Llama-2 and Gemma, for retrieval-based tasks like Retrieval Augmented Generation (RAG). It takes your existing large language model and custom dataset as input, and outputs a fine-tuned model optimized for specific downstream applications. This is ideal for those working with extensive prompt sequences who need to adapt LLMs without consuming excessive GPU memory.

No commits in the last 6 months.

Use this if you need to fine-tune large language models for RAG or other retrieval-based tasks and are constrained by GPU memory or desire significantly faster training times.

Not ideal if you are looking for a general-purpose LLM training library that doesn't focus on tensor-parallelism or specific memory optimization for retrieval tasks, or if you prefer a PyTorch-only ecosystem.

large-language-models retrieval-augmented-generation model-fine-tuning machine-learning-engineering natural-language-processing
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 7 / 25
Maturity 16 / 25
Community 3 / 25

How are scores calculated?

Stars

35

Forks

1

Language

Python

License

Category

llm-fine-tuning

Last pushed

Apr 25, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/transformers/aniquetahir/JORA"

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