DeepSpeed alternatives for AI agents.

The Triton Inference Server provides an optimized cloud and edge inferencing solution.

Tensors and Dynamic neural networks in Python with strong GPU acceleration

Run, manage, and scale AI workloads on any AI infrastructure. Use one system to access & manage all AI compute (Kubernetes, Slurm, 20+ clouds, on-prem).

Open Machine Learning Compiler Framework

Time series forecasting with PyTorch

A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit and 4-bit floating point (FP8 and FP4) precision on Hopper, Ada and Blackwell GPUs, to provide better performance with lower memory utilization in both training and inference.

On-device AI across mobile, embedded and edge for PyTorch

The fastai deep learning library

🤗 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.

Ray is an AI compute engine. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.

Pretrain, finetune ANY AI model of ANY size on 1 or 10,000+ GPUs with zero code changes.

Deep Learning for humans

Ultralytics YOLOv5 in PyTorch > ONNX > CoreML > TFLite

Deepfakes Software For All

🧑‍🏫 60+ Implementations/tutorials of deep learning papers with side-by-side notes 📝; including transformers (original, xl, switch, feedback, vit, ...), optimizers (adam, adabelief, sophia, ...), gans(cyclegan, stylegan2, ...), 🎮 reinforcement learning (ppo, dqn), capsnet, distillation, ... 🧠

We write your reusable computer vision tools. 💜