Open source repo

What's inside ray-project/ray

ray-project/ray is a Python repo with 1,514,135 lines written in 2,949 files. ray is built with FastAPI and uses Redis, MongoDB, Google, and Google Cloud Storage. AI signed 3.3% of the 32,244 commits of ray.

Its description on GitHub: “Ray is an AI compute engine. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.”

PythonApache-2.044,168

Read on October 31, 2026, at commit 45c4123.

What the card proved: what it’s built with on the left, what it uses on the right.

See each one with its proof

Its folders

TechnologiesWhat is ray built with?ray is built with FastAPI and uses Redis, MongoDB, Google, and Google Cloud Storage.

Who writes itHow much of ray did AI write?3.3%AI signed 3.3% of the 32,244 commits of ray.

In figures

lines written: 1,514,135; The ones people and agents wrote: product, tests, schemas and tooling. Generated code doesn’t count.
files: 2,949; The product’s, not counting tests.
test files: 2,145; The files that test the code.
technologies: 5; Each one with the file and line that prove it.

SeeRepo read its code facts-only: no AI, and none of its code is kept.

Frequently asked questions

What is ray-project/ray?

ray-project/ray is a Python repo with 1,514,135 lines written in 2,949 files. Its description on GitHub: “Ray is an AI compute engine. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.”

The proof: its page on GitHub

What is ray built with?

ray is built with FastAPI and uses Redis, MongoDB, Google, and Google Cloud Storage.

The proof: each technology with its file and line

How much of ray did AI write?

AI signed 3.3% of the 32,244 commits of ray.

The proof: the commits each tool signed

What license does ray have?

ray is released under the Apache-2.0 license.

The proof: its page on GitHub

How many stars does ray have on GitHub?

ray had 44,168 stars on GitHub on October 31, 2026.