lodestone-rock/RamTorch

Memory Management for the GPU Poor, run the latest open source frontier models on consumer Nvidia GPUs

🚀 A simple way to launch, train, and use PyTorch models on almost any device and distributed configuration, automatic mixed precision (including fp8), and easy-to-configure FSDP and DeepSpeed support

AirLLM 2.8T LLM inference with single 4GB GPU

Supercharge Your Model Training

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

High-throughput tensor loading for PyTorch

A PyTorch Extension: Tools for easy mixed precision and distributed training in Pytorch

Making large AI models cheaper, faster and more accessible

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