# facebookarchive/fbcunn

Facebook's extensions to torch/cunn.

Repository: https://github.com/facebookarchive/fbcunn
Canonical: https://ross.abutalabs.com/products/fbcunn
Language: Lua
License: NOASSERTION
License Family: other
Archived: true
Last push: 2017-07-21T22:04:03+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": 4284, "days_push": 3330, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, archived, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1059, forks 242 (observed 2026-08-28T04:03:25.461922+00:00)

## What it is
Facebook AI Research's CUDA-accelerated deep learning extension modules for the Torch-7 framework, including fast FFT-based convolutions, multi-GPU parallel containers, and optimized lookup tables. It is an archived repository and no longer maintained.

## Use cases
- accelerate convolutional neural network training on GPUs
- train deep learning models across multiple GPUs
- build fast neural language models with efficient lookup tables
- classify a very large number of classes with hierarchical softmax
- speed up temporal convolutions in Torch-7

## When to choose
- you are already working in the legacy Torch-7 ecosystem and need faster GPU modules
- you need the original FFT-based convolution implementations from the FAIR paper

## When to avoid
- you are starting a new project - use PyTorch or another maintained framework instead
- you need up-to-date GPU support or bug fixes
- you do not use Torch-7 or Lua

## Facets
- artifact type: library
- maturity: abandoned
- function: deep-learning, machine-learning, gpu-computing
- domain: deep-learning, machine-learning, computer-vision
- platform: lua
- tags: torch7, cuda, fft-convolution, facebook-ai-research, neural-networks, multi-gpu, archived, natural-language-processing, gpu, linux

## Member repositories
- facebookarchive/fbcunn (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:25.461922+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 2026-08-30T06:57:13.902059+00:00, confidence not recorded.
  - readme: https://github.com/facebookarchive/fbcunn (fetched 2026-08-28T04:03:25.461922+00:00, sha 1c1b83a58a7b)
- Data as of 2026-08-30T08:39:29.467469+00:00.