OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Search & Research AI

Nanosage

Nanosage is an open-source Python command-line research assistant that combines local documents and web results in a knowledge base, recursively explores relevant subqueries, and generates structured Markdown reports.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Nanosage — OpenViST project profile in Search & Research AI.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Nanosage when a research question benefits from exploring related subtopics across local files and web results, and you want the search structure reflected in a generated report.

Who it is for

  • Researchers who want to investigate a topic across a local document collection and web sources.
  • Python users seeking a command-line workflow that records search branches and produces Markdown reports.

Documented uses

  • Search a local corpus or combine it with web retrieval for a research question.
  • Expand a query into relevant subqueries and track the exploration in a table of contents.
  • Generate a synthesized Markdown report from retrieved material.
Before you choose

Requirements & limits

The project is distributed as a public GitHub repository and identified as open source under the Apache-2.0 license. Its documented interface is a Python CLI for local/self-hosted use.

  • Web search uses Tavily and requires an API key; local-documents-only operation is also documented.
  • Retrieval models serve different inputs: SigLIP and CLIP support vision and text, ColPali is listed for documents, and all-MiniLM for text-focused retrieval.
  • The --max_depth option controls subquery recursion and is documented with a default of one.
  • Python 3.8 or later and Ollama with Gemma 2B are included in the quick-start prerequisites.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up Nanosage for local-model research

Prepare the documented Python, Ollama/Gemma and Tavily configuration before running a research query.

Before you start

  • Python 3.8 or later
  • Project requirements file
  • Ollama and the Gemma 2B model
  • Tavily API key for web retrieval

Steps

  1. Install the project dependencies with the command below.

    Documented command
    pip install -r requirements.txt

  2. Install Ollama and pull the gemma2:2b model.

  3. Copy env.example to .env and add the Tavily API key.

  4. Optionally copy config.example.yaml to config.yaml for runtime defaults.

Expected outcome

The documented dependencies and configuration are prepared for a subsequent Nanosage run.

Guide limitations

  • This guide covers setup only; it does not verify that the installation succeeds.
  • The Tavily key is needed for web retrieval; the README also documents local-documents-only runs.
Documentation 1 ↗
Source-based practical guide

Run a query against local documents and web sources

Use the documented CLI pattern to search a local corpus alongside web results.

Before you start

  • Nanosage dependencies and runtime configuration
  • A local document directory
  • A Tavily API key for web retrieval

Steps

  1. Choose a natural-language research query and a directory containing supported documents.

  2. Run the command below.

    Documented command
    python main.py --query "quantum computing" --corpus_dir ./my_documents --retrieval_model siglip --web_search

  3. Review the Markdown report in the query-specific directory under results/.

Expected outcome

The documented workflow combines local and web retrieval and writes a detailed Markdown report.

Guide limitations

  • Web retrieval requires Tavily configuration.
  • The listed corpus formats are .txt, .md, .py, .json, .yaml, .csv, PDF and common image formats.
  • The CLI documents a default recursion depth of one.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
12 October 2026
Original website title
masterFoad/NanoSage · Official homepage title · retained in its source language
How we review sourcesSuggest a correction