SwarmCoder
Overview How it works Agents Principles Status About
APACHE 2.0 GITHUB ↗ ● IN DEVELOPMENT
UNDER ACTIVE DEVELOPMENT · NOT YET RELEASED

Where this is today

This site describes a system that is substantially built — not one you can download yet. The code will be published under Apache 2.0 in the coming weeks.

WHAT EXISTS
✓ The requirement and story model, with executable checks
✓ Document intake and requirement extraction
✓ The fixed workflows and their machine-checked invariants
✓ The swarm scheduler and worker sandboxing
✓ The verification harness and candidate archival
✓ Full session tracing and persistence
✓ The operator console described on these pages
WHAT IS HONESTLY NOT PROVEN YET

The full pipeline has been driven by tests and simulated model servers far more than by live ones. The claim that a swarm of small local models beats a single frontier pass on real work in a real repository is the thesis this project exists to test.

It is well supported by published research and not yet by our own numbers. When we have those numbers, they will be on this page — including the ones that disappoint.

It runs on two machines

TOKEN SERVER
A DGX Spark-class box serving tokens and nothing else. Any OpenAI-compatible endpoint works for the local roles — the worker model is a configuration choice, deliberately interchangeable.
ORCHESTRATOR
An ordinary workstation running the orchestrator, the sandboxes and the builds — on your network, against your repository.

What SwarmCoder is not

Not an IDE plugin
A standalone system with its own console. Chat is a companion — available everywhere, required nowhere.
Not configurable into a different process
No plugins, no agent graphs, no workflow DSL. That is a feature — and it is the feature.
Not fast
Built for long unsupervised runs — start it in the evening, judge the deliveries in the morning.
Not a hosted service
It runs on your hardware, on your network, on your repository.
Not a way to avoid deciding
It asks you exactly twice per story, and it expects real answers.
What it is
One process, followed every time, with the receipts. See the cycle →
APACHE 2.0
Open source when it ships — the same license as ZeroZ4j, the framework it is built on.
GITHUB · COMING SOON
The repository link will appear here, and in the header of every page, when the code is published.

If you want an agent that will do anything you ask in any order, there are excellent ones. SwarmCoder is for the case where you would rather have one process, followed every time, with the receipts.

Questions, or want to be notified? franzschoning.com ↗
SwarmCoder

Agentic software development with full traceability — built on ZeroZ4j by Franz Schöning, Principal Enterprise Architect.

swarmcoder.dev GitHub ↗ soon zeroz4j.com ↗ franzschoning.com ↗ ● In development · Apache 2.0
© 2026 Franz Schöning. Released under the Apache License 2.0.