BeatIt

Local-first oncology case research — ingest your library, run sourced assessments, and discuss treatment options in AI Chat.

View on GitHub Release notes

What it does

BeatIt is a FastAPI web app for organizing oncology research material and synthesizing insights with explicit source citations. It runs locally with SQLite storage, supports Ollama or OpenRouter for LLM calls, and optional secured deployment on Render.

Library ingest — notes, PDFs, URLs, YouTube transcripts, video, DICOM imaging
Home assessment — baseline synthesis with executive summary, open items, and numbered references
AI Chat — multi-turn treatment-options discussion scoped to your library
Chat observations — pin excerpts from chat into the next assessment or save to library
Imaging vision — optional slice-level reads via Ollama vision models
PDF export — download assessments with reference appendix

Quick start

git clone https://github.com/simonbright/beatit.git
cd beatit
cp .env.example .env
python3 -m venv .venv && source .venv/bin/activate
pip install -r requirements.txt
python run.py

Open http://localhost:8080. See the README for OpenRouter, Ollama, and Tailscale VM setup.

Changelog

v1.0.0 — August 2026

First public release: library ingest, baseline assessment, custom tasks, AI Chat, chat observations for analysis updates, imaging vision, PDF export, Render deploy blueprint.

All releases on GitHub ↗

Medical disclaimer: BeatIt supports research and case organization. It is not medical advice and does not replace evaluation by qualified oncology teams.