# microsoft/CNTK

Microsoft Cognitive Toolkit (CNTK), an open source deep-learning toolkit

Repository: https://github.com/microsoft/CNTK
Canonical: https://ross.abutalabs.com/products/cntk
Homepage: https://docs.microsoft.com/cognitive-toolkit/
Language: C++
License: NOASSERTION
License Family: other
Topics: cognitive-toolkit, cntk, deep-learning, machine-learning, deep-neural-networks, neural-network, distributed, python, c-plus-plus, c-sharp, java
Archived: true
Last push: 2023-03-11T07:31:35+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 3933, "days_push": 1271, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: 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 17595, forks 4222 (observed 2026-08-28T04:11:20.042166+00:00)

## What it is
The Microsoft Cognitive Toolkit (CNTK) is an open-source deep-learning toolkit that describes neural networks as directed computational graphs, supporting DNNs, CNNs, and RNNs/LSTMs with automatic differentiation and distributed multi-GPU/multi-server training. It is usable as a library from Python, C#, C++, and Java, or standalone via its BrainScript model description language.

## Use cases
- train deep neural networks across multiple gpus and servers
- build cnn and rnn/lstm models in python
- embed a deep learning library in c++ or c# applications
- evaluate trained neural network models from java
- run distributed sgd training with automatic differentiation

## When to choose
- you must maintain or extend an existing CNTK-based codebase
- you need a legacy deep-learning toolkit for 64-bit Windows or Linux

## When to avoid
- starting any new deep-learning project - CNTK is no longer actively developed
- you want modern framework support, ecosystem, and community updates (use PyTorch or TensorFlow instead)

## Facets
- artifact type: library
- maturity: abandoned
- function: deep-learning, machine-learning, gpu-computing
- domain: deep-learning, machine-learning, artificial-intelligence
- platform: windows, python, cpp, cross-platform
- tags: neural-networks, automatic-differentiation, distributed-training, brainscript, microsoft, discontinued, linux

## Member repositories
- microsoft/CNTK (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:20.042166+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 2026-08-29T17:02:46.009171+00:00, confidence not recorded.
  - readme: https://github.com/microsoft/CNTK (fetched 2026-08-28T04:11:20.042166+00:00, sha 5561599ad481)
  - homepage: https://docs.microsoft.com/cognitive-toolkit/ (fetched 2026-08-29T08:00:50.867979+00:00, sha d3f99ab2f812)
  - site_page: https://learn.microsoft.com/en-us/lifecycle/faq/internet-explorer-microsoft-edge (fetched 2026-08-29T08:00:50.870453+00:00, sha c363b2f43d4b)
- Data as of 2026-08-30T08:39:29.467469+00:00.