OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Developer Tools & SDKs

Peft Singlora

PEFT-SingLoRA is a Python library that implements SingLoRA, a single-matrix low-rank adaptation method for parameter-efficient transformer fine-tuning. It registers the adapter with Hugging Face PEFT so developers can configure targeted linear layers in Transformers workflows.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Peft Singlora — OpenViST project profile in AI Developer Tools & SDKs.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when adapting a PEFT-based Transformers fine-tuning workflow to test or use single-matrix low-rank adapters.

Who it is for

  • Developers maintaining model fine-tuning workflows with Hugging Face PEFT and Transformers.
  • Practitioners looking to reduce adapter parameter counts compared with standard LoRA.

Documented uses

  • Register SingLoRA as an adapter option in a PEFT workflow.
  • Configure targeted model layers with a familiar PEFT LoRA configuration.
  • Track SingLoRA training steps in a loop using gradient accumulation.
Before you choose

Requirements & limits

The repository is public and licensed under BSD 2-Clause.

  • The project reports half the adapter parameters of standard LoRA at the same rank.
  • The README specifies Python 3.9+, PyTorch 1.13+, PEFT 0.7+, and Transformers 4.34+.
  • The documented workflow registers SingLoRA before using PEFT's get_peft_model.
  • Quantized base-model compatibility and adapter merging or unmerging are roadmap items.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Register SingLoRA in a PEFT fine-tuning workflow

Configure a Transformers model with SingLoRA adapters through PEFT.

Before you start

  • Python 3.9 or newer
  • PyTorch 1.13 or newer, on CPU or CUDA
  • PEFT 0.7 or newer
  • Transformers 4.34 or newer

Steps

  1. Install the package with the command below.

    Documented command
    pip install peft-singlora

  2. Load a model with Transformers and call setup_singlora() to register the adapter.

  3. Create a PEFT LoraConfig with a rank and target modules, then pass the model and configuration to get_peft_model.

Expected outcome

A PEFT model configured to use SingLoRA for the targeted linear layers.

Guide limitations

  • The README's quick-start snippet is a simplified training-loop example and does not specify supported model names.
  • Quantized base-model compatibility is listed as a roadmap item.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
14 October 2026
Original website title
revaldianggara/PEFT-SingLoRA · Official homepage title · retained in its source language
How we review sourcesSuggest a correction