← The openness index

classic-ml · Openness profile

XGBoost

DMLC (Apache Software Foundation project) · snapshot 2026-11-07

Scalable, portable and open-source gradient boosting library

Tier 5 Fully open / reproducible

What the publisher claims: open source

The one-liner: Fully open by construction: the artifact is the method.

The six questions

A AccessScore 3 of 3

Full source and binaries are downloadable.

P PermissionScore 3 of 3

Apache 2.0.

D DeploymentScore 3 of 3

Runs on CPU, GPU and distributed backends with no proprietary runtime.

T TransparencyScore 3 of 3

Algorithm, objective, hyperparameters and code are fully public, and the data is the user's own, so there is no hidden corpus to disclose.

R ReproducibilityScore 3 of 3

A skilled group can rebuild an equivalent implementation from the published algorithm and code.

G Governance & ParticipationScore 3 of 3

Community-governed dmlc project with distributed maintainership and a steady stream of merged outside PRs.

License

Apache 2.0 OSI-approved — license source ↗

What is actually released

Why it matters

The baseline the open-model debate is measured against; in classical ML, open is the default rather than a concession.

Sources

Derived from the openness dataset behind But Is It Open? — snapshot 2026-11-07 ↗. Scores are qualitative readings of the cited evidence, not verdicts; tier measures openness, not intelligence.

← The openness indexJoin the newsletter ↗