OSS Tanbou
View on GitHubProject website

LightGBM — fast and memory-efficient gradient-boosted trees at scale

OSS scale score 307.1OSS health 92Update pace 19 commits/30d
About these scores

OSS scale score is an unbounded metric that log-compresses and weights Stars, Watchers, Forks, and Contributors. Discovery score is the current OSS scale score minus the score at discovery. Update pace is commits in the last 30 days, growth momentum is the OSS scale score difference within the recent observation window, and OSS health is a 0–100 rating based on available recency, Community Health, and release data.

Stars
18,781
Primary language
C++
License
MIT
Repository last updated
Sep 17, 2026

Overview

LightGBM is a tree-based gradient-boosting framework designed for fast, memory-efficient training. It supports classification, regression, ranking, parallel and distributed execution, and GPU learning, making it a common alternative to other boosted-tree libraries for structured data.

Features and best fit

Based on official documentation; not hands-on tested · Content checked:

Train gradient-boosted trees efficiently

LightGBM is designed to reduce training time and memory use while providing high-performance tree-based gradient boosting for large datasets.

Sources: [2][4]

Scale across GPU, parallel, and distributed training

The project supports GPU acceleration and distributed learning in addition to single-machine workflows, with interfaces including Python and R.

Sources: [2]

For classification, regression, and ranking on structured data

LightGBM is a strong candidate when comparing boosted trees with linear models, random forests, and other methods on tabular datasets.

Sources: [2][4]

Tune model complexity and update old repository references

Depth, leaves, learning rate, and boosting rounds materially affect accuracy and overfitting. The README also notes that the official repository moved from Microsoft/LightGBM to the current organization in March 2026.

Sources: [2]

Official sources

  1. [1]lightgbm-org/LightGBM repository(2026-09-20)
  2. [2]LightGBM README(2026-09-20)
  3. [3]LightGBM MIT license(2026-09-20)
  4. [4]LightGBM documentation(2026-09-20)
Supplemental curator note

Like XGBoost, LightGBM is a strong choice for structured data, with an emphasis on fast and memory-efficient training plus GPU and distributed modes. The official repository moved from `Microsoft/LightGBM` to `lightgbm-org/LightGBM` in March 2026, so older links and automation may need updating.

Try it in 3 steps

  1. 1

    Install LightGBM

    Install the released LightGBM package in a virtual environment for the Python API.

    python -m pip install lightgbm
  2. 2

    Train a tiny boosted-tree model

    Use a tiny binary-classification dataset to verify the basic train/predict flow.

    python -c "import lightgbm as lgb, numpy as np; X=np.array([[0],[1],[2],[3]]); y=np.array([0,0,1,1]); m=lgb.train({'objective':'binary','verbose':-1}, lgb.Dataset(X,label=y), num_boost_round=5); print(m.predict([[1.5]]))"
  3. 3

    Add validation and tuning

    For real evaluation, separate validation data and tune learning rate, leaves, and boosting rounds while monitoring overfitting.

    Add valid_sets + early_stopping callback
Check the official README

Growth

Growth trends · Last 30 days

18,781 Stars

Trend data is still being collected.

Development activity

Last 90 days · weekly

Commits (last 30 days)
19
Open PRs
71
Issues opened
25
Issues closed
21
PRs opened
71
PRs merged
47

Issues

25 / 21

Jun 23Sep 20
Issues openedIssues closed

Pull requests

71 / 47

Jun 23Sep 20
PRs openedPRs merged

Maintenance

Median first response
2.1 days
Issue response rate
40% (4/10)

Based on up to the 100 newest issues opened by external users in the last 90 days. A first comment from an OWNER, MEMBER, or COLLABORATOR counts as a response; issues whose full comment history cannot be checked are excluded. The median and response rate update weekly.

Built with

Categories and tags

Categories

Tags

GitHub data

GitHub dataView detailed GitHub data
View on GitHub

GitHub Topics

  • gbdt
  • gbm
  • machine-learning
  • data-mining
  • distributed
  • lightgbm
  • gbrt
  • decision-trees
  • gradient-boosting
  • python
  • r
  • parallel
Stars
18,781
Forks
4,071
Watchers
422
Open issues
447
Contributors
341
Primary language
C++
License
MIT
Repository last updated
Sep 17, 2026
Write a related article

Share a guide or use case for this OSS in Markdown. Articles are published after administrator approval.

Report incorrect information

Tell us if any listing information is incorrect or outdated.