Project AURIONAI Use-Cases Research & Innovation

TabArena

Elo across tabular tasks, anchored RandomForest default = 1000; TabPFN-lab researchers are involved

All Taskstabularactive

Score in Elo, All Tasks, no imputation, snapshot 2026-05-06, 59 entries, tuned + ensembled

  1. 1Prior LabsRealTabPFN-2.51,604 Elomaintainer-run
  2. 2INRIARealMLP1,514 Elomaintainer-run
  3. 3Layer 6 AITabDPT1,460 Elomaintainer-run
20scores
13models
0independent
20maintainer-run
0vendor-reported

Definition

Version
All Tasks
Category
tabular
Maintainer
TabArena team
Status
active
Home page
https://benchmarklist.com/benchmarks/tabarena_all/

Sources

  1. TabArena All Tasks Benchmark Scores & AI Model Leaderboard | BenchmarkList benchmarklist.com/benchmarks/tabarena_all https://benchmarklist.com/benchmarks/tabarena_all/
  2. The state of Tabular Foundation Models (2026) mindfulmodeler.substack.com https://mindfulmodeler.substack.com/p/the-state-of-tabular-foundation-models
  3. TabArena All Tasks Benchmark Scores & AI Model Leaderboard | BenchmarkList benchmarklist.com/benchmarks/tabarena_all https://benchmarklist.com/benchmarks/tabarena_all/
  4. tabarena.ai https://tabarena.ai/

Scores at a Glance

Score in Elo, All Tasks, no imputation, snapshot 2026-05-06, 59 entries, tuned + ensembled

Higher is better for this unit. Elo has no zero, so the axis starts at 1,000.

1Prior LabsRealTabPFN-2.5Prior Labs
1,604 Elo
maintainer-run2INRIARealMLPINRIA
1,514 Elo
maintainer-run3Layer 6 AITabDPTLayer 6 AI
1,460 Elo
maintainer-run4YandexTabMYandex
1,448 Elo
maintainer-run5Microsoft AILightGBMMicrosoft AI
1,435 Elo
maintainer-run6YandexCatBoostYandex
1,418 Elo
maintainer-run7DMLC communityXGBoostDMLC community
1,376 Elo
maintainer-run8RandomForest
1,172 Elo
maintainer-run
1,0001,2001,4001,600

Score in Elo, All Tasks, no imputation, snapshot 2026-05-06, 59 entries, default

Higher is better for this unit. Elo has no zero, so the axis starts at 1,000.

1Prior LabsTabPFN-2.6Prior Labs
1,626 Elo
maintainer-run2INRIATabICL v2INRIA
1,598 Elo
maintainer-run3YandexCatBoostYandex
1,368 Elo
maintainer-run4SAPSAP-RPT-OSS (sap-rpt-1-oss)SAP
1,284 Elo
maintainer-run5DMLC communityXGBoostDMLC community
1,214 Elo
maintainer-run6Microsoft AILightGBMMicrosoft AI
1,188 Elo
maintainer-run
1,0001,2001,4001,600

Score in Elo, All Tasks, no imputation, snapshot 2026-05-06, 59 entries, tuned

Higher is better for this unit. Elo has no zero, so the axis starts at 1,330.

1YandexCatBoostYandex
1,407 Elo
maintainer-run2Microsoft AILightGBMMicrosoft AI
1,387 Elo
maintainer-run3DMLC communityXGBoostDMLC community
1,352 Elo
maintainer-run
1,3301,3501,3701,3901,410
Where this comes from
Each bar is one model, with its most independent recorded score. Hatched bars are vendor-reported; the word beside each bar says who ran it. Settings and dates differ between rows; the Scores table below has every result.

Scores

20 of 20 scores shown tabarena-scores.csv
ModelScoreWho ran itSettingView or populationDateSource
AmazonAutoGluon 1.51,700 Elomaintainer-runextreme, 4hrank 1 of 59, All Tasks, no imputation, snapshot 2026-05-06, 59 entries6 May 2026benchmarklist.com
Prior LabsTabPFN-2.61,626 Elomaintainer-rundefaultrank 2 of 59, All Tasks, no imputation, snapshot 2026-05-06, 59 entries6 May 2026benchmarklist.com
Prior LabsRealTabPFN-2.51,604 Elomaintainer-runtuned + ensembledrank 3 of 59, All Tasks, no imputation, snapshot 2026-05-06, 59 entries6 May 2026benchmarklist.com
INRIATabICL v21,598 Elomaintainer-rundefaultrank 4 of 59, All Tasks, no imputation, snapshot 2026-05-06, 59 entries6 May 2026benchmarklist.com
INRIARealMLP1,514 Elomaintainer-runtuned + ensembledrank 8 of 59, All Tasks, no imputation, snapshot 2026-05-06, 59 entries6 May 2026benchmarklist.com
Layer 6 AITabDPT1,460 Elomaintainer-runtuned + ensembledrank 9 of 59, All Tasks, no imputation, snapshot 2026-05-06, 59 entries6 May 2026benchmarklist.com
YandexTabM1,448 Elomaintainer-runtuned + ensembledrank 10 of 59, All Tasks, no imputation, snapshot 2026-05-06, 59 entries6 May 2026benchmarklist.com
Microsoft AILightGBM1,435 Elomaintainer-runtuned + ensembledrank 11 of 59, All Tasks, no imputation, snapshot 2026-05-06, 59 entries6 May 2026benchmarklist.com
YandexCatBoost1,418 Elomaintainer-runtuned + ensembledrank 13 of 59, All Tasks, no imputation, snapshot 2026-05-06, 59 entries6 May 2026benchmarklist.com
YandexCatBoost1,407 Elomaintainer-runtunedrank 14 of 59, All Tasks, no imputation, snapshot 2026-05-06, 59 entries6 May 2026benchmarklist.com
Microsoft AILightGBM1,387 Elomaintainer-runtunedrank 18 of 59, All Tasks, no imputation, snapshot 2026-05-06, 59 entries6 May 2026benchmarklist.com
DMLC communityXGBoost1,376 Elomaintainer-runtuned + ensembledrank 19 of 59, All Tasks, no imputation, snapshot 2026-05-06, 59 entries6 May 2026benchmarklist.com
YandexCatBoost1,368 Elomaintainer-rundefaultrank 20 of 59, All Tasks, no imputation, snapshot 2026-05-06, 59 entries6 May 2026benchmarklist.com
DMLC communityXGBoost1,352 Elomaintainer-runtunedrank 22 of 59, All Tasks, no imputation, snapshot 2026-05-06, 59 entries6 May 2026benchmarklist.com
SAPSAP-RPT-OSS (sap-rpt-1-oss)1,284 Elomaintainer-rundefaultrank 28 of 59, All Tasks, no imputation, snapshot 2026-05-06, 59 entries6 May 2026benchmarklist.com
DMLC communityXGBoost1,214 Elomaintainer-rundefaultrank 33 of 59, All Tasks, no imputation, snapshot 2026-05-06, 59 entries6 May 2026benchmarklist.com
Microsoft AILightGBM1,188 Elomaintainer-rundefaultrank 38 of 59, All Tasks, no imputation, snapshot 2026-05-06, 59 entries6 May 2026benchmarklist.com
RandomForest1,172 Elomaintainer-runtuned + ensembledrank 39 of 59, All Tasks, no imputation, snapshot 2026-05-06, 59 entries6 May 2026benchmarklist.com
PerpetualBooster1,086 Elomaintainer-runbestrank 45 of 59, All Tasks, no imputation, snapshot 2026-05-06, 59 entries6 May 2026benchmarklist.com
RandomForest1,000 Elomaintainer-rundefault (Elo anchor)All Tasks, no imputation, snapshot 2026-05-06, 59 entries6 May 2026benchmarklist.com
Where this comes from
Scores from different settings, harnesses or dates are not directly comparable; the row says which. Sorted by score, highest first, as a reading aid, not a ranking.

To cite this page: Project AURION, AI Use-Cases Research & Innovation, "TabArena", https://aurion.semai.id/benchmarks/tabarena/, records of October 2026.