# dmlc/xgboost

Scalable, Portable and Distributed Gradient Boosting (GBDT, GBRT or GBM) Library,  for Python, R, Java, Scala, C and more. Runs on single machine, Hadoop, Spark, Dask, Flink and DataFlow

Repository: https://github.com/dmlc/xgboost
Canonical: https://ross.abutalabs.com/products/xgboost
Homepage: https://xgboost.readthedocs.io/
Language: C++
License: Apache-2.0
License Family: permissive
Topics: gbdt, gbrt, gbm, distributed-systems, xgboost, machine-learning
Last push: 2026-08-26T10:53:24+00:00

## Health v2 (maintenance only)
Score: 95/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 99, release rhythm 86, longevity 100
- inputs: {"age_days": 4591, "days_push": 7, "days_rel": 18, "gap_med": 34.5, "n_releases_24m": 17}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 28694, forks 8890 (observed 2026-08-28T04:11:52.759070+00:00)

## What it is
XGBoost is an optimized, scalable gradient boosting library implementing parallel tree boosting (GBDT/GBM) with bindings for Python, R, Java, Scala, C++, and more. It runs on a single machine or distributed environments such as Spark, Dask, Kubernetes, and Ray, with GPU support and the ability to handle billions of examples.

## Use cases
- train gradient boosted tree models on tabular data
- win kaggle-style classification and regression problems
- train XGBoost models on a Spark or Dask cluster
- accelerate model training with NVIDIA GPUs
- build a ranking model with learning to rank
- handle categorical features and missing values in tree models
- serve fast predictions for structured data

## When to choose
- you need state-of-the-art accuracy on structured/tabular data
- your dataset is too large for one machine and needs distributed training
- you want mature multi-language bindings and GPU acceleration
- you need battle-tested, widely supported ML tooling

## When to avoid
- you are working with unstructured data like images, audio, or raw text where deep learning excels
- you need very low-latency, tiny models for embedded deployment
- you want simple interpretable linear models
- you need online/incremental learning as a primary workflow

## Facets
- artifact type: library
- maturity: stable
- function: machine-learning, gpu-computing, data-science
- domain: machine-learning, data-science, microservices
- platform: python, jvm, cpp, cross-platform, cloud
- tags: gradient-boosting, gbdt, tree-ensembles, tabular-data, spark, dask, scikit-learn, gpu

## Member repositories
- dmlc/xgboost (main) score 95

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:52.759070+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 2026-08-29T16:54:20.478631+00:00, confidence not recorded.
  - readme: https://github.com/dmlc/xgboost (fetched 2026-08-28T04:11:52.759070+00:00, sha 2960b9ad23af)
  - homepage: https://xgboost.readthedocs.io/ (fetched 2026-08-29T07:50:38.819672+00:00, sha 55c1b5a84793)
- Data as of 2026-08-30T08:39:29.467469+00:00.