6 projects
petals
Easy way to efficiently run 100B+ language models without high-end GPUs
test-petals
Easy way to efficiently run 100B+ language models without high-end GPUs
hivemind
Decentralized deep learning in PyTorch
prefetch-generator
a simple tool to compute arbitrary generator in a background thread
lean-transformer
PyTorch transformers that don't hog your GPU memory.
as-a-service
a simple tool to turn your function into a background service