Local Deep Research
A self-hosted research assistant that runs multi-step searches with local or cloud language models.
These buttons open the developer's own site, repository or store listing in a new tab. wares.gg does not host downloads.
About Local Deep Research
Local Deep Research breaks a question into several research steps, searches sources and writes up the findings. It supports local models through llama.cpp and Ollama as well as cloud providers such as Google, and can query more than ten search engines, including arXiv and PubMed, alongside your own private documents.
Data is kept locally and encrypted. It is deployed with Docker, with Unraid templates provided, and the project is under very active development. It suits researchers who want deep-research style reports without sending everything to a hosted service.
Strengths
- Works with local models or cloud providers
- Searches arXiv, PubMed and more than ten other sources
- Can include your own private documents
- Local, encrypted storage
Limitations
- Good results with local models need a capable GPU
- Self-hosting setup is more involved than a hosted service
Details
- Pricing
- FreeFree, with source code published on GitHub; cloud model providers may charge.
- License
- Open source, license not stated
- Developer
- LearningCircuit
- Platforms
- Self-hosted
- How it runs
- Self-hosted
- Best suited for
- Researchers who want private, multi-step research reports using their own models
- Categories
- Local AI tools, AI assistants
- Last verified
- Added
- Provenance
- Facts checked against the developer's own pages and store listings, 1 sources on file.
Alternatives to Local Deep Research
Compare allSoftware that can replace Local Deep Research for an important use case, and what changes if you switch.
Vane (formerly Perplexica)
Self-hosted, privacy-focused AI search engine with cited answers.
Vane is an MIT self-hosted AI search engine with cited answers and local Ollama support, focused on web search rather than multi-step reports.
Khoj
Open-source AI second brain that researches across your documents and the web with any model.
Khoj is an AGPL-3.0 self-hosted assistant that researches across your documents and the web, adding scheduled automations and custom agents.
Open Notebook
Self-hosted, open source alternative to Google's NotebookLM.
Open Notebook is an MIT self-hosted NotebookLM alternative centred on your own sources rather than searching arXiv and PubMed.
Similar software
Related functionality, not necessarily a direct replacement.
RAGFlow
An open-source, self-hostable RAG engine with document ingestion, hybrid search and a visual agent builder.
Kotaemon
An open-source, self-hosted RAG app for chatting with your documents using local or hosted models.
AnythingLLM
A desktop AI app that answers questions about your own documents, with no account and no API key needed.
Ollama
The simplest way to pull down an open language model and run it on your own machine, from one command or a desktop app.
Docling
Open source library that converts documents into AI-ready structured formats.