🤗 PEFT: State-of-the-art Parameter-Efficient Fine-Tuning.

The simplest, fastest repository for training/finetuning medium-sized GPTs.

20+ high-performance LLMs with recipes to pretrain, finetune and deploy at scale.

Easily turn large sets of image urls to an image dataset. Can download, resize and package 100M urls in 20h on one machine.

Automatic Speech Recognition with Speaker Diarization based on OpenAI Whisper

Merge Transformers language models by use of gradient parameters.

Fine tune SDXL on YouTube videos

LlamaIndex is the leading framework for building LLM-powered agents over your data.

Set of Utilities I Have Coded to Help Me Train RPGv6 on Flux1

Supercharge Your Model Training