google-research/tabfm — GitHub trend

FINDING #8666 · UNIT ID 1271618207

google-research/tabfm

TabFM (Tabular Foundation Model) is a pretrained tabular foundation model developed by Google Research for tabular data regression and classification.

[ PYTHON ][ ORG ][ GITHUB ↗ ]

SURPRISE SCORE

0.00

Score Breakdown

SURPRISE0.0000554

ENGAGEMENT0.48

FRESHNESS1.00

SCORE = SURPRISE × ENGAGEMENT^0.7 × FRESHNESS × VISIBILITY × CONFIDENCE

SURPRISE = WINDOW STARS / DAYS / (AUDIENCE + FLOOR)

1% OF STARS IN ARCHIVE

Growth Telemetry

VELOCITY /D

4.86

ACCEL

+0.04

RETENTION

64.3%

PEAK 2026-08-26 · FORK-RETENTION 16.7% · 34 STARS / WINDOW

Author Audience

AUDIENCE

87,621

FOLLOWERS

17,096

OWNER ★

267,147

Engagement Signals

FORKS

263

ISSUE AUTH

PR AUTH

UNIQUE STARGAZERS 34 / 34 (DIVERSITY 1.00)

Why This Is A Finding

google-research/tabfm собрал 34 звёзд за окно, тогда как у автора всего 17,096 подписчиков — эффективная аудитория ≈ 87,621. Это даёт surprise-индекс 0.0000554 (звёзды относительно охвата автора, а не в абсолюте). Удержание форков 16.7% и 0 внешних контрибьюторов отделяют реальный инструмент от разовой вспышки. Акселерация положительная — рост ещё не выдохся.

Related Findings

RANKS ABOVE 11% OF 9760 FINDINGS

#08MddIdd/mdd-sim-gateway[ PYTHON ]0.41 #17litaolemo/xtquant_big_convert[ PYTHON ]0.36 #20LBH-123-AI/Comfyui_Minimax_h3_latent_Upscaler[ PYTHON ]0.34 #24zeenie-ai/OpenCompany[ PYTHON ]0.30 #25917Dhj/DeepPaperNote[ PYTHON ]0.29

METRICS IN CONTEXT

MEDIAN ACROSS ALL 9760 FINDINGS · Δ vs MEDIAN · PERCENTILE = SHARE RANKED BELOW

METRICVALUEMEDIANΔ MEDPERCENTILE

SCORE0.000.00-0.00ABOVE 11%

VELOCITY4.863.71+1.14ABOVE 59%

RETENTION64.3%38.0%+26.3 PPABOVE 85%

FORKS26386+177ABOVE 76%

SURPRISE0.000.01-0.01ABOVE 3%