Train transformer language models with reinforcement learning.

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

🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.

Active Learning for Text Classification in Python

Accessible large language models via k-bit quantization for PyTorch.

A collection of LogitsProcessors to customize and enhance LLM behavior for specific tasks.

EasyR1: An Efficient, Scalable, Multi-Modality RL Training Framework based on veRL

Convert PDF to markdown + JSON quickly with high accuracy

Our solution for the arc challenge 2024

Quick implementation of nGPT, learning entirely on the hypersphere, from NvidiaAI

PyTorch native post-training library

CLIP (Contrastive Language-Image Pretraining), Predict the most relevant text snippet given an image

SigLIP-based Aesthetic Score Predictor

SVG Differentiable Rendering: Generating vector graphics using neural networks. Support: text-to-SVG, Image-to-SVG, SVG Editing.

Differentiable Vector Graphics Rasterization

[CVPR 2024] Official implementation for "SVGDreamer: Text Guided SVG Generation with Diffusion Model" https://arxiv.org/abs/2312.16476

StarVector is a foundation model for SVG generation that transforms vectorization into a code generation task. Using a vision-language modeling architecture, StarVector processes both visual and te…

Scour - An SVG Optimizer / Cleaner

convert pixels to SVG square-based shapes

Fully open reproduction of DeepSeek-R1

Leverage pyppeteer and the imagetrace.js library to trace a bitmap to SVG in python

A SOTA lightweight multilingual LLM

Python package to easily retrain OpenAI's GPT-2 text-generating model on new texts