byerlikaya/septum repository overview

⁠Your data never leaves. Your AI still works.

The all-in-one image. Everything Septum needs in a single container: FastAPI backend, Next.js dashboard, SQLite by default, no external services required. Simplest way to try Septum.

⁠Quick start

docker run --name septum \
  --add-host=host.docker.internal:host-gateway \
  -p 3000:3000 \
  -v septum-data:/app/data \
  -v septum-uploads:/app/uploads \
  -v septum-anon-maps:/app/anon_maps \
  -v septum-vector-indexes:/app/vector_indexes \
  -v septum-bm25-indexes:/app/bm25_indexes \
  -v septum-models:/app/models \
  byerlikaya/septum

Open http://localhost:3000⁠ — the setup wizard picks database, cache, LLM provider, and regulations. No .env file needed.

⁠Tags

Use -gpu if you have an NVIDIA GPU available to the container — batch document ingestion, OCR, and Whisper transcription all become measurably faster.

⁠Volumes

⁠Looking for the modular layout?

This all-in-one image bundles everything. For air-gap deployments, split zones, or custom orchestration, see the sibling images: septum-api⁠, septum-web⁠, septum-gateway⁠, septum-audit⁠, septum-mcp⁠.

⁠Links

• Source: github.com/byerlikaya/Septum⁠
• Architecture: docs/ARCHITECTURE.md⁠
• Benchmark: docs/BENCHMARK.md⁠
• Issues: github.com/byerlikaya/Septum/issues⁠