LearningCircuit/local-deep-research — GitHub trend

FINDING #78 · UNIT ID 929868765

LearningCircuit/local-deep-research

~95% on SimpleQA (e.g. Qwen3.6-27B on a 3090). Supports all local and cloud LLMs (llama.cpp, Ollama, Google, ...). 10+ search engines - arXiv, PubMed, your private documents. Everything Local & Encrypted.

[ PYTHON ][ GITHUB ↗ ]

SURPRISE SCORE

0.00

Score Breakdown

SURPRISE0.00435

ENGAGEMENT1.29

FRESHNESS1.11

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

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

5% OF STARS IN ARCHIVE

Growth Telemetry

VELOCITY /D

4.63

ACCEL

-0.23

RETENTION

3.2%

PEAK 2026-05-09 · FORK-RETENTION 36.0% · 417 STARS / WINDOW

Author Audience

AUDIENCE

1,026

FOLLOWERS

145

OWNER ★

8,805

Engagement Signals

FORKS

775

ISSUE AUTH

PR AUTH

UNIQUE STARGAZERS 417 / 417 (DIVERSITY 1.00)

Why This Is A Finding

LearningCircuit/local-deep-research собрал 417 звёзд за окно, тогда как у автора всего 145 подписчиков — эффективная аудитория ≈ 1,026. Это даёт surprise-индекс 0.00435 (звёзды относительно охвата автора, а не в абсолюте). Удержание форков 36.0% и 122 внешних контрибьюторов отделяют реальный инструмент от разовой вспышки. Акселерация отрицательная — внимание остывает после пика.

Related Findings

RANKS ABOVE 91% OF 891 FINDINGS

#07noonghunna/club-3090[ PYTHON ]0.0148 #131bananachicken/MaaNTE[ PYTHON ]0.012 #15denuitt1/mhr-cfw[ PYTHON ]0.0115 #23debpalash/OmniVoice-Studio[ PYTHON ]0.0105 #31mattzh72/articraft[ PYTHON ]0.00963

METRICS IN CONTEXT

MEDIAN ACROSS ALL 891 FINDINGS · Δ vs MEDIAN · PERCENTILE = SHARE RANKED BELOW

METRICVALUEMEDIANΔ MEDPERCENTILE

SCORE0.010.00+0.01ABOVE 91%

VELOCITY4.632.86+1.78ABOVE 72%

RETENTION3.2%7.6%-4.4 PPABOVE 16%

FORKS7752,034-1,259ABOVE 29%

SURPRISE0.000.00+0.00ABOVE 93%