# lightgbm-org/LightGBM

A fast, distributed, high performance gradient boosting (GBT, GBDT, GBRT, GBM or MART) framework based on decision tree algorithms, used for ranking, classification and many other machine learning tasks.

Repository: https://github.com/lightgbm-org/LightGBM
Canonical: https://ross.abutalabs.com/products/lightgbm
Homepage: https://lightgbm.readthedocs.io/en/latest/
Language: C++
License: MIT
License Family: permissive
Topics: gbdt, gbm, machine-learning, data-mining, distributed, lightgbm, gbrt, decision-trees, gradient-boosting, python, r, parallel, kaggle
Last push: 2026-08-26T03:44:51+00:00

## Health v2 (maintenance only)
Score: 86/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 61, longevity 100
- inputs: {"age_days": 3680, "days_push": 7, "days_rel": 46, "gap_med": 518, "n_releases_24m": 2}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 18714, forks 4057 (observed 2026-08-28T04:11:27.050765+00:00)

## What it is
LightGBM is a fast, distributed, high-performance gradient boosting framework based on decision tree algorithms, with APIs for Python, R, C, and other languages. It supports parallel, distributed, and GPU learning and is designed for large-scale data with low memory usage.

## Use cases
- train gradient boosted decision tree models for classification
- build ranking models for search and recommendation
- fit models on large datasets with distributed training
- speed up model training with GPU learning
- compete in tabular machine learning competitions like Kaggle
- tune GBDT hyperparameters for better accuracy

## When to choose
- you need fast, memory-efficient gradient boosting on tabular data
- your dataset is too large for a single machine and you need distributed training
- you want GPU-accelerated tree learning
- you need mature Python, R, or C APIs for boosting

## When to avoid
- you need deep learning on unstructured data like images or text
- you want a simple, interpretable linear model
- your project requires a different ecosystem like Spark MLlib exclusively

## Facets
- artifact type: library
- maturity: stable
- function: machine-learning, data-science, gpu-computing
- domain: machine-learning, data-science, big-data
- platform: python, cpp, cross-platform, windows
- tags: gradient-boosting, gbdt, decision-trees, distributed-training, ranking, classification, regression, kaggle, gpu, linux, macos

## Member repositories
- lightgbm-org/LightGBM (main) score 86

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:27.050765+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 2026-08-29T17:01:19.659647+00:00, confidence not recorded.
  - readme: https://github.com/lightgbm-org/LightGBM (fetched 2026-08-28T04:11:27.050765+00:00, sha 22d1546c92b2)
  - homepage: https://lightgbm.readthedocs.io/en/latest/ (fetched 2026-08-29T07:59:17.854214+00:00, sha 046846998c9e)
- Data as of 2026-08-30T08:39:29.467469+00:00.