tabular
LightGBM
Gradient boosting library; best GBDT on TabArena after tuning and ensembling.
Levelmodel
Classtabular
Parametersnot in the notes
Contextnot in the notes
Released18 Jul 2026
Opennessopen source
Runs onCPU server
Statusactive
Benchmark Results
maintainer-run
TabArena All Taskstabular
tuned + ensembledmaintainer-run
1,435
8001,3001,800
Score in Elo; the axis starts at 800, as Elo has no zero. Higher is better.
rank 11 of 59, All Tasks, no imputation, snapshot 2026-05-06, 59 entries
tunedmaintainer-run
1,387
8001,3001,800
Score in Elo; the axis starts at 800, as Elo has no zero. Higher is better.
rank 18 of 59, All Tasks, no imputation, snapshot 2026-05-06, 59 entries
defaultmaintainer-run
1,188
8001,3001,800
Score in Elo; the axis starts at 800, as Elo has no zero. Higher is better.
rank 38 of 59, All Tasks, no imputation, snapshot 2026-05-06, 59 entries
Where this comes from
Every score names who ran it: vendor-reported, maintainer-run or independent. A rank is shown with the population it was ranked in (gap rule 3).
Identity
- Developer
- Microsoft AI
- Family
- not in the notes
- Released
- 18 Jul 2026 sourcesregistry date
- Weights published
- –
- Record checked
- 10 Oct 2026
Where this comes from
Developer and dates from the research notes. Release date is the vendor's public announcement; weights date is when open weights were published.
Architecture and Size
- Architecture
- gradient boosting
- Total parameters
- – sources
- Active parameters
- –
- Checkpoint parameters
- –
- Context length
- not in the notes
- Input
- table
- Output
- table
Where this comes from
Parameter counts as the developer states them. Where the checkpoint holds a different total, both are shown (register A16).
Openness and Access
- Openness
- open source
- Licence
- MIT License sources
- Access routes
- open weights download
Where this comes from
The licence file governs over registry tags and aggregator labels.
Deployment
- Runs on
- CPU server
- Footprint
- Latest 4.7.0 on PyPI 18 Jul 2026
- Runtimes
- not in the notes
Where this comes from
Deployment tiers and runtimes as named in the notes. Footprint is the notes' figure, not a measurement by AURION.
Indonesia
- Indonesian support
- not stated
- Regional languages
- not in the notes
- Jakarta hosting
- unknown
- Languages
- not in the notes
Where this comes from
Indonesia fields are carried on every model; unknown is an allowed value.
Lineage
The notes record no lineage for this model.
Industry Use Cases
No industry use case links this model.
Links
DeveloperMicrosoft AIbig tech, USA
Microsoft AIbig tech, USAModelLightGBMMicrosoft AI, tabular
LightGBMMicrosoft AI, tabularUse CasesOil Palm Yield PredictionAgricultureRoute OptimisationRetail and LogisticsUse cases of models in the same R&D topic; none is recorded for this model itself.
Who Runs It
PT Astra Agro Lestari TbkIn production, 21 Sep 2026
J&T Express IndonesiaIn production, 26 Aug 2026Unnamed Indonesian estate (15,000 to 17,000 ha)In production, 2024Deployments of these use cases; none names this model.
PT Astra Agro Lestari TbkIn production, 21 Sep 2026J&T Express IndonesiaIn production, 26 Aug 2026Unnamed Indonesian estate (15,000 to 17,000 ha)In production, 2024Deployments of these use cases; none names this model.Organisations 1
Sources
- Releases , lightgbm-org/LightGBM github.com https://github.com/microsoft/LightGBM/releases
- pypi.org/pypi/lightgbm/json https://pypi.org/pypi/lightgbm/json
- GitHub - lightgbm-org/LightGBM: A fast, distributed, high performance gradient boosting (GBT, GBDT, GBRT, GBM or MART) framework based on de github.com https://github.com/microsoft/LightGBM
To cite this page: Project AURION, AI Use-Cases Research & Innovation, "LightGBM", https://aurion.semai.id/models/microsoft-lightgbm/, records of October 2026.