Library/LoRAForked

microsoft/LoRA

LoRA

Code for loralib, an implementation of "LoRA: Low-Rank Adaptation of Large Language Models"

View on GitHub↗Upstream microsoft/LoRA↗

Builder

Microsoft

microsoft • big-tech

Stars

13,385

Using upstream star count

Forks

894

Using upstream fork count

Open Issues

Activity Score

0/100

0 commits in 30d

Created

Jun 18, 2021

Project creation date

README Summary

LoRAlib is Microsoft's implementation of Low-Rank Adaptation (LoRA), a technique for efficiently fine-tuning large language models by learning low-rank updates to weight matrices. The library provides PyTorch modules that can replace standard neural network layers to enable parameter-efficient training with significantly reduced memory requirements. LoRA achieves comparable performance to full fine-tuning while using orders of magnitude fewer trainable parameters.

AI Dev Skills

Unmapped

LoRA Fine-tuningParameter-Efficient Fine-tuningLow-Rank Matrix AdaptationLarge Language Model OptimizationNeural Network Weight DecompositionTransfer Learning

Taxonomy

AI Trends

Parameter-Efficient Fine-tuning Large Language Models Model Compression

Deployment Context

Self-hosted Cloud API On-premise

Modalities

Text

Skill Areas

LoRA Fine-tuning Parameter-Efficient Fine-tuning Low-Rank Matrix Adaptation Large Language Model Optimization Neural Network Weight Decomposition Transfer Learning

Use Cases

Custom Language Model Fine-tuning Domain-Specific Model Adaptation Multi-task Model Training Resource-Constrained Model Customization

Recent Activity

Updated 1 years ago

7 Days

30 Days

90 Days

Quality

research

Quality: medium
Maturity: research

PM Skills

Scale & Reliability

Languages

Python100.0%

Timeline

Project created: Jun 18, 2021
Forked: Mar 13, 2026
Your last push: 1 years ago
Upstream last push: 1 years ago
Tracked since: Dec 17, 2024

Similar Repos

pgvector cosine similarity · $0

Loading…