Stone to Silicon

Artificial intelligence · 2016

PyTorch

An eager, Pythonic deep learning framework that researchers adopt en masse.

Facebook AI Research released PyTorch in 2016, building on Lua Torch with dynamic computation graphs. It became the dominant research framework and underlies most open language model work.

See it in the living treeWikipedia

Arcs above join it to what it stood on, arcs below to what it led to. The bar places this stretch on the tree's whole timeline.

What it built on 2

Related measurements

The dashed line marks when it first worked.

Training compute of notable AI models Source: Epoch AI, Notable AI Models (CC BY 4.0)

ML hardware FLOP/s per dollar Source: Epoch AI, ML hardware (CC BY 4.0)