# tensorflow/tensor2tensor

Library of deep learning models and datasets designed to make deep learning more accessible and accelerate ML research.

Repository: https://github.com/tensorflow/tensor2tensor
Canonical: https://ross.abutalabs.com/products/tensor2tensor
Language: Python
License: Apache-2.0
License Family: permissive
Topics: machine-learning, machine-translation, deep-learning, reinforcement-learning, tpu
Archived: true
Last push: 2023-06-02T18:55:09+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 3366, "days_push": 1188, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: archived
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 17464, forks 3783 (observed 2026-08-28T04:11:19.897184+00:00)

## What it is
Tensor2Tensor (T2T) is a Python library of deep learning models and datasets built on TensorFlow, developed by the Google Brain team to make deep learning research more accessible. It is now deprecated in favor of its successor library Trax, though it continues to receive bug fixes.

## Use cases
- train a transformer model for machine translation
- run image classification experiments on MNIST
- train language models and summarization models
- experiment with deep learning models on TPUs
- benchmark sequence-to-sequence models
- learn deep learning through runnable notebooks

## When to choose
- you need the original Transformer implementation or T2T model zoo for reproducing research papers
- you want a one-command training pipeline with bundled datasets and hyperparameter sets
- you are maintaining legacy research code built on T2T

## When to avoid
- you are starting a new project - use Trax, JAX, or PyTorch instead
- you need actively maintained models or modern hardware support
- you want production deployment rather than research experimentation

## Facets
- artifact type: library
- maturity: maintenance
- function: deep-learning, machine-learning, llm-training, speech-recognition, nlp
- domain: deep-learning, machine-learning, computer-vision, speech-processing
- platform: python
- tags: tensorflow, tpu, transformer, deprecated, research, datasets, model-zoo, natural-language-processing, gpu, linux, macos

## Member repositories
- tensorflow/tensor2tensor (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:19.897184+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 2026-08-29T17:02:55.913057+00:00, confidence not recorded.
  - readme: https://github.com/tensorflow/tensor2tensor (fetched 2026-08-28T04:11:19.897184+00:00, sha a2c9b7e9dde1)
  - registry_pypi: https://pypi.org/pypi/tensor2tensor/json (fetched 2026-08-29T08:00:58.802236+00:00, sha b0a95b2557b7)
- Data as of 2026-08-30T08:39:29.467469+00:00.