DeepYard
A

Axolotl

2

Streamlined tool for fine-tuning large language models

Open Sourcefree

About

Axolotl is a popular open-source tool designed to streamline the fine-tuning of large language models. It supports a wide range of model architectures and training techniques including LoRA, QLoRA, full fine-tuning, RLHF, and DPO. Configuration is done through simple YAML files, making it accessible for both researchers and practitioners to customize LLMs on their own data.

Details

Typefine-tuning-tool
LanguagesPython
Supported ModelsLlama, Mistral, Mixtral, Falcon, GPT-NeoX, Phi, Qwen, Gemma

Tags

fine-tuningloraqlorarlhfdpollmyaml-configopen-source