# 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.

Repository: https://github.com/LearningCircuit/local-deep-research
Canonical: https://ross.abutalabs.com/products/local-deep-research
Language: Python
License: MIT
License Family: permissive
Topics: academia, arxiv, brave, deep-research, local, local-llm, mistral, pubmed, research, research-tool, retrieval-augmented-generation, searxng, self-hosted, local-deep-research, home-automation, homeserver, ollama, anthropic, openai, encryption
Last push: 2026-08-26T22:48:41+00:00

## Health v2 (maintenance only)
Score: 83/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 99, release rhythm 86, longevity 40
- inputs: {"age_days": 570, "days_push": 7, "days_rel": 17, "gap_med": 1.0, "n_releases_24m": 169}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 8995, forks 793 (observed 2026-08-28T04:10:26.684142+00:00)

## What it is
A self-hosted AI research assistant that performs deep, iterative multi-source research using local or cloud LLMs (Ollama, llama.cpp, OpenAI, Anthropic, etc.) and 10+ search engines including arXiv, PubMed, and private document collections. All data is stored locally with SQLCipher encryption, achieving ~95% accuracy on SimpleQA benchmarks.

## Use cases
- run deep research queries entirely on local LLMs without sending data to the cloud
- search arXiv and PubMed for academic literature with AI synthesis
- ask questions over my private documents with citations
- build a self-hosted alternative to paid deep research tools
- compare answers across multiple search engines and LLM providers
- research on an air-gapped or privacy-sensitive machine

## When to choose
- you need privacy-first research with everything local and encrypted
- you want to combine local LLMs with many search backends including private document collections
- you want a self-hosted web UI plus Docker deployment for iterative research workflows

## When to avoid
- you need a lightweight single-shot Q&A without iterative search overhead
- you have no GPU or local model and don't want to rely on cloud LLM APIs
- you need a hosted managed service with zero setup

## Facets
- artifact type: application
- maturity: active
- function: rag, search-engine, llm-inference, web-scraping, chatbot
- domain: large-language-models, self-hosted, artificial-intelligence
- platform: python, self-hosted, cross-platform
- tags: deep-research, local-llm, ollama, searxng, encrypted, research-assistant, arxiv, pubmed, private-documents, retrieval-augmented-generation, search, docker, web-server

## Member repositories
- LearningCircuit/local-deep-research (main) score 83

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:26.684142+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:24:22.262395+00:00, confidence not recorded.
  - readme: https://github.com/LearningCircuit/local-deep-research (fetched 2026-08-28T04:10:26.684142+00:00, sha 21d8159597b7)
  - registry_pypi: https://pypi.org/pypi/local-deep-research/json (fetched 2026-08-29T08:24:22.252627+00:00, sha 717e8b9ceb85)
- Data as of 2026-08-30T08:39:29.467469+00:00.