# microsoft/DMTK

Microsoft Distributed Machine Learning Toolkit

Repository: https://github.com/microsoft/DMTK
Canonical: https://ross.abutalabs.com/products/dmtk
Homepage: http://www.dmtk.io
License: MIT
License Family: permissive
Topics: dmtk, multiverso, lightgbm, microsoft, machine-learning
Archived: true
Last push: 2018-09-12T21:14:58+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": 3951, "days_push": 2912, "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 2738, forks 554 (observed 2026-08-28T04:07:16.388231+00:00)

## What it is
DMTK is Microsoft's Distributed Machine Learning Toolkit, an umbrella project hosting a parameter server framework (Multiverso) plus distributed ML applications like LightLDA, distributed word embedding, and LightGBM. It enables large-scale distributed training of topic models, embeddings, and gradient boosting trees.

## Use cases
- train machine learning models across a cluster
- run distributed gradient boosting for classification and ranking
- scale topic modeling to large corpora
- compute word embeddings in parallel
- build custom distributed ML algorithms on a parameter server

## When to choose
- you need a parameter server framework for custom distributed ML training
- you want a historical reference implementation of distributed topic modeling or word embedding

## When to avoid
- you need actively maintained distributed training tooling
- you only need LightGBM, which lives in its own actively developed repository
- you want modern deep learning training infrastructure

## Facets
- artifact type: framework
- maturity: abandoned
- function: machine-learning, llm-training
- domain: machine-learning, microservices, big-data
- platform: windows, cross-platform
- tags: parameter-server, distributed-training, gradient-boosting, topic-modeling, word-embedding, microsoft, linux

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

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:16.388231+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 2026-08-30T08:19:13.670686+00:00, confidence not recorded.
  - readme: https://github.com/microsoft/DMTK (fetched 2026-08-28T04:07:16.388231+00:00, sha 920313367303)
- Data as of 2026-08-30T08:39:29.467469+00:00.