# tensorflow/swift

Swift for TensorFlow

Repository: https://github.com/tensorflow/swift
Canonical: https://ross.abutalabs.com/products/tensorflow-swift
Homepage: https://tensorflow.org/swift
Language: Jupyter Notebook
License: Apache-2.0
License Family: permissive
Topics: machine-learning, differentiable-programming, tensorflow
Archived: true
Last push: 2022-01-12T19:55:30+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 3053, "days_push": 1694, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, archived
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 6122, forks 610 (observed 2026-08-28T04:09:36.172636+00:00)

## What it is
Swift for TensorFlow was an experimental next-generation machine learning platform integrating differentiable programming directly into the Swift language. The project was archived in February 2021 and the repository is now read-only, though its ideas live on in the Swift compiler and spin-off projects.

## Use cases
- train neural networks in Swift
- differentiable programming with gradients in Swift
- run deep learning models on GPU from Swift
- use Swift in Jupyter notebooks for ML
- learn how ML and compiler design intersect

## When to choose
- you want to study archived research on language-integrated differentiable programming
- you need the historical Swift deep learning API or model garden
- you're researching how differentiable programming was added to Swift

## When to avoid
- you need a maintained ML framework for production
- you want ecosystem support, updates, or bug fixes
- you're starting a new machine learning project - use PyTorch, JAX, or MLX instead

## Facets
- artifact type: library
- maturity: abandoned
- function: machine-learning, deep-learning, compiler, interpreter
- domain: machine-learning, deep-learning, compilers, programming-languages
- platform: -
- tags: differentiable-programming, tensorflow, archived, jupyter-notebooks, language-integration, swift, linux, macos, gpu

## Member repositories
- tensorflow/swift (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:36.172636+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:48:00.214195+00:00, confidence not recorded.
  - readme: https://github.com/tensorflow/swift (fetched 2026-08-28T04:09:36.172636+00:00, sha 1317528a759e)
  - homepage: https://tensorflow.org/swift (fetched 2026-08-29T08:44:36.710568+00:00, sha 6e6193c05a3b)
- Data as of 2026-08-30T08:39:29.467469+00:00.