# fastai/fastai

The fastai deep learning library

Repository: https://github.com/fastai/fastai
Canonical: https://ross.abutalabs.com/products/fastai
Homepage: http://docs.fast.ai
Language: Jupyter Notebook
License: Apache-2.0
License Family: permissive
Topics: deep-learning, machine-learning, pytorch, python, gpu, fastai, notebooks, colab
Last push: 2026-08-15T09:18:08+00:00

## Health v2 (maintenance only)
Score: 93/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 97, release rhythm 83, longevity 100
- inputs: {"age_days": 3280, "days_push": 18, "days_rel": 34, "gap_med": 60, "n_releases_24m": 10}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 28125, forks 7641 (observed 2026-08-28T04:11:48.126012+00:00)

## What it is
fastai is a deep learning library built on PyTorch that offers high-level components for quickly achieving state-of-the-art results in vision, text, tabular, and collaborative filtering tasks, plus low-level APIs for research. It follows a layered architecture and is accompanied by a free course and book for learning deep learning.

## Use cases
- train an image classifier in a few lines of code
- fine-tune a text sentiment model
- build a tabular model on structured data
- create a recommendation system
- do image segmentation
- learn deep learning with a practical course
- prototype neural network models on GPU in Colab

## When to choose
- you want fast, high-level PyTorch training with sensible defaults
- you are learning deep learning and want a gentle but powerful API
- you need quick state-of-the-art baselines for vision, text, or tabular data
- you want a mix of high-level convenience and low-level extensibility for research

## When to avoid
- you need full low-level control over every training detail from the start
- your project requires frameworks other than PyTorch
- you need production serving infrastructure, which fastai does not provide
- you prefer writing raw PyTorch training loops

## Facets
- artifact type: library
- maturity: stable
- function: machine-learning, deep-learning, image-processing, nlp, data-science
- domain: deep-learning, machine-learning, data-science, computer-vision
- platform: python, cross-platform, cli
- tags: pytorch, training-api, notebooks, colab, transfer-learning, tabular-data, computer-vision, education, natural-language-processing, gpu

## Member repositories
- fastai/fastai (main) score 93

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:48.126012+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 2026-08-29T16:54:27.027701+00:00, confidence not recorded.
  - readme: https://github.com/fastai/fastai (fetched 2026-08-28T04:11:48.126012+00:00, sha b86713e085fa)
  - homepage: http://docs.fast.ai (fetched 2026-08-29T07:51:08.661579+00:00, sha d63b7dec5b2e)
  - registry_pypi: https://pypi.org/pypi/fastai/json (fetched 2026-08-29T07:51:08.670918+00:00, sha 3e22f19f6efa)
- Data as of 2026-08-30T08:39:29.467469+00:00.