Trending Projects

Extensible, parallel implementations of the popular t-SNE dimensionality reduction algorithm for data visualization.

Official implementation of a controllable character video synthesis model using spatial decomposed modeling.

WikiChat is an improved Retrieval Augmented Generation (RAG) model that reduces language model hallucination by retrieving data from a corpus.

A video conversation model that combines LLM capabilities with pretrained visual encoders for video-based chatbots.

A collective communications library with various primitives for multi-machine training.

A developer-friendly natural language processing library for building chatbots, extracting entities, and analyzing sentiment.

Properly implemented ResNet-s for CIFAR10/100 in PyTorch, matching the original paper description.

A unified deep learning model for correspondence, stereo, and depth estimation tasks.

A repository of Azure OpenAI samples to complement the OpenAI cookbook.

A machine learning project walkthrough in Python demonstrating the end-to-end ML pipeline on a real-world dataset.

Fine-tune and deploy the Whisper speech recognition model with accelerated inference and support for various platforms.

A reusable, open-source library for building AI-powered Go game agents in a readable manner.

A powerful and easy-to-use toolkit for implementing various computer vision tasks on Android, including text recognition, barcode scanning, image labeling, face detection, and object detection.

A modular voice assistant app for experimenting with state-of-the-art transcription, response generation, and text-to-speech models.

A Python toolbox for class-incremental learning, a machine learning technique for training models on a continuous stream of data.

A Python library for generating 3D worlds using simulation of plates, rain shadow, erosion, and more.

An open-source platform for logging and analyzing OpenAI API usage, including cost tracking and prompt improvement.

An open-source platform for building and deploying ML pipelines with a focus on MLOps

SynapseML is a simple and distributed machine learning library for building and deploying AI models at scale.

A curated collection of papers and resources related to various aspects of face recognition technology.

Probabilistic reasoning and statistical analysis tools built on TensorFlow for data scientists and ML researchers.

ECMAScript proposal for type syntax that is erased, focused on vibe coders using AI tools.

The official DeiT repository for a PyTorch-based vision transformer model for image classification.

PyTorch implementation of popular reinforcement learning algorithms like A2C, PPO, ACKTR and GAIL for developers.

An Android app that demonstrates a ChatGPT-powered chatbot using OpenAI's API and the Stream Chat SDK for Compose.

Interactive tool to understand the behavior of machine learning models in an extensible, framework-agnostic way.

A Python library for human pose estimation, skeleton-based action recognition, and action synthesis.

Transform ML models into native code (Java, C, Python, Go, etc.) with zero dependencies

Turn expensive prompts into cheap fine-tuned models for AI developers and ML engineers.

DirectML is a high-performance, hardware-accelerated DirectX 12 library for machine learning tasks on a variety of GPUs.

An open-source web client that enables users to create videos from text using OpenAI's Sora model.

A powerful BERT-based language model for natural language understanding and representation learning.

A PyTorch pre-trained model for real-time interest point detection, description, and sparse tracking.

A high-performance C++ library for upscaling anime videos using AI-powered super-resolution techniques.

SAM is a PyTorch-based library that implements Sharpness-Aware Minimization, an optimization algorithm for training more robust deep learning models.

Minimal, clean example of LSTM neural network training in Python for learning purposes.

A real-time 3D talking face generation library for developers building AI-powered virtual assistants and avatars.

AIdea is a powerful app that supports GPT, Chinese language models, Stable Diffusion, and various AI features.

A repository of 60 useful data science prompts for ChatGPT to help developers build with AI tools.

An easy-to-use Python library for processing and manipulating 3D point clouds and meshes.

A Python library that generates collaborative AI agents for complex tasks using GPT models.

Unified-Modal Speech-Text Pre-Training for Spoken Language Processing

A Python SDK/API for reverse-engineered Google Bard, a powerful AI chatbot.

Real-time facial emotion detection using deep learning and computer vision.

PyTorch implementation of state-of-the-art reinforcement learning algorithms for vibe coders.

A PyTorch implementation of the Quasi-Recurrent Neural Network, up to 16x faster than cuDNN LSTM.

A reference BLEU implementation that auto-downloads test sets and reports a version string.

A repository that provides daily updates on the latest research papers related to large language models (LLMs).

A pure Python implementation of Spiking Neural Networks for tasks like MNIST classification.

A Python framework for comprehensive diagnosis and optimization of AI agents using simulated, realistic synthetic interactions.