https://github.com/NVIDIA/workbench-llamafactory

This is an NVIDIA AI Workbench example project that demonstrates an end-to-end model development workflow using Llamafactory.
https://github.com/NVIDIA/workbench-llamafactory

Last synced: about 2 months ago
JSON representation

This is an NVIDIA AI Workbench example project that demonstrates an end-to-end model development workflow using Llamafactory.

• Host: GitHub
• URL: https://github.com/NVIDIA/workbench-llamafactory
• Owner: NVIDIA
• License: apache-2.0
• Created: 2024-05-31T20:07:30.000Z (5 months ago)
• Default Branch: main
• Last Pushed: 2024-09-17T09:35:45.000Z (about 2 months ago)
• Last Synced: 2024-09-17T12:10:52.054Z (about 2 months ago)
• Language: Jupyter Notebook
• Size: 273 KB
• Stars: 28
• Watchers: 2
• Forks: 8
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

• awesome-LLM-resourses - workbench-llamafactory - to-end model development workflow using Llamafactory. (微调 Fine-Tuning)
• alan_awesome_llm - workbench-llamafactory - to-end model development workflow using Llamafactory. (微调 Fine-Tuning)
• alan_awesome_llm - workbench-llamafactory - to-end model development workflow using Llamafactory. (微调 Fine-Tuning)

README

        
## LLaMA-Factory - AI Workbench Project

This is an [NVIDIA AI Workbench](https://www.nvidia.com/en-us/deep-learning-ai/solutions/data-science/workbench/) project to deploy [LLaMA-Factory](https://github.com/hiyouga/LLaMA-Factory).

**Please follow the LLM fine-tuning tutorial for [RTX AI Toolkit here](https://github.com/NVIDIA/RTX-AI-Toolkit/blob/main/tutorial-llama3-finetune.md).**



Minimum recommended Workbench ver: 0.50.16

Docker Desktop ver. 4.31+

## Features

- Fine-tune an LLM using Llama factory.

- Fuse generated LoRA adapter with the base model.

- Apply Post Training Quantization to your model.

- Export model to TensorRT-LLM checkpoint.

- Support for unsloth.