Stars
Optimized primitives for collective multi-GPU communication
RAFT contains fundamental widely-used algorithms and primitives for machine learning and information retrieval. The algorithms are CUDA-accelerated and form building blocks for more easily writing …
Run your own AI cluster at home with everyday devices 📱💻 🖥️⌚
Hydra is a framework for elegantly configuring complex applications
Progressive delivery Kubernetes operator (Canary, A/B Testing and Blue/Green deployments)
A toolkit to run Ray applications on Kubernetes
Implement a ChatGPT-like LLM in PyTorch from scratch, step by step
A fancy self-hosted monitoring tool
End-to-end stack for WebRTC. SFU media server and SDKs.
A high-throughput and memory-efficient inference and serving engine for LLMs
All the goodies from both Mixpanel and Plausible combined into one tool.
Course to get into Large Language Models (LLMs) with roadmaps and Colab notebooks.
Code release for NeRF (Neural Radiance Fields)
Ray is an AI compute engine. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.
StreamDiffusion: A Pipeline-Level Solution for Real-Time Interactive Generation
Gorilla: Training and Evaluating LLMs for Function Calls (Tool Calls)
Documentation and source code powering Twitter's Community Notes
Hackable and optimized Transformers building blocks, supporting a composable construction.
Open-sourced codes for MiniGPT-4 and MiniGPT-v2 (https://minigpt-4.github.io, https://minigpt-v2.github.io/)
Chat with your database or your datalake (SQL, CSV, parquet). PandasAI makes data analysis conversational using LLMs and RAG.
JARVIS, a system to connect LLMs with ML community. Paper: https://arxiv.org/pdf/2303.17580.pdf
🤗 PEFT: State-of-the-art Parameter-Efficient Fine-Tuning.
Train transformer language models with reinforcement learning.