google-research/timesfm — GitHub trend

FINDING #7636 · UNIT ID 793784055

google-research/timesfm

TimesFM (Time Series Foundation Model) is a pretrained time-series foundation model developed by Google Research for time-series forecasting.

[ PYTHON ][ ORG ][ GITHUB ↗ ]

SURPRISE SCORE

0.00

Score Breakdown

SURPRISE0.00036

ENGAGEMENT0.41

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

31.57

ACCEL

-4.82

RETENTION

23.1%

PEAK 2026-08-24 · FORK-RETENTION 55.2% · 221 STARS / WINDOW

Author Audience

AUDIENCE

87,603

FOLLOWERS

17,096

OWNER ★

267,189

Engagement Signals

FORKS

2,768

ISSUE AUTH

PR AUTH

UNIQUE STARGAZERS 221 / 221 (DIVERSITY 1.00)

Why This Is A Finding

google-research/timesfm собрал 221 звёзд за окно, тогда как у автора всего 17,096 подписчиков — эффективная аудитория ≈ 87,603. Это даёт surprise-индекс 0.00036 (звёзды относительно охвата автора, а не в абсолюте). Удержание форков 55.2% и 0 внешних контрибьюторов отделяют реальный инструмент от разовой вспышки. Акселерация отрицательная — внимание остывает после пика.

Related Findings

RANKS ABOVE 22% 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 22%

VELOCITY31.573.71+27.86ABOVE 92%

RETENTION23.1%38.0%-14.9 PPABOVE 29%

FORKS2,76886+2,682ABOVE 97%

SURPRISE0.000.01-0.01ABOVE 9%