torchtune

PyTorch native post-training library

View on GitHub↗Upstream meta-pytorch/torchtune↗

Builder

meta-pytorch

meta-pytorch • individual

Stars

5,716

Using upstream star count

Forks

712

Using upstream fork count

Open Issues

Activity Score

0/100

0 commits in 30d

Created

Oct 20, 2023

Project creation date

README Summary

TorchTune is a PyTorch-native library for easily authoring, fine-tuning and experimenting with LLMs. It provides a modular design with interchangeable components for training recipes, model architectures, tokenizers, and datasets, supporting both full fine-tuning and memory-efficient techniques like LoRA and QLoRA.

AI Dev Skills

Unmapped

Tags

LoRA Fine-tuningQLoRA Parameter-Efficient TrainingFull Fine-tuningLarge Language Model TrainingPyTorch Model OptimizationMulti-GPU TrainingMemory-Efficient TrainingModel QuantizationGradient CheckpointingMixed Precision TrainingDistributed TrainingCloud TrainingCustom Language Model Fine-tuningModel Performance OptimizationParameter-Efficient Fine-tuningEfficient AI TrainingNatural Language ProcessingCode Generation Model TrainingResearch ExperimentationSelf-hostedOpen Source Language ModelsResearchDeveloper ToolsInstruction TuningTextCodeModel CustomizationDomain-Specific Model AdaptationMulti-GPU ClustersChat Model TrainingDemocratized AIPython

Recent Activity

Updated 1 months ago

7 Days

30 Days

90 Days

Quality

beta

Quality: high
Maturity: beta

PM Skills

Developer Platform

Languages

Python100.0%

Timeline

Project created: Oct 20, 2023
Forked: Mar 13, 2026
Your last push: 1 months ago
Upstream last push: 7 days ago
Tracked since: Mar 12, 2026

Similar Repos

pgvector cosine similarity · $0

Loading…