Our products
Both products start from the same idea: AI should not depend on one machine or one place.
Dune
Distributed LLM inference for critical AI infrastructure
Dune pools machines an organization already owns into one cluster that serves language models through an OpenAI-compatible API.
It spreads the work across ordinary machines, with CPUs or GPUs, and keeps running when some of them are lost.
Key features
- No single point of failure Distributed AI inference that removes the reliance on a single machine, a single cluster or a single data center.
- Self-healing Peer-to-peer AI inference that self-heals when nodes fail, so critical AI systems always stay on.
- Answers, not errors When part of the system fails, people still get their answers.
- Safe model updates New model versions roll out carefully and roll back on their own if they perform worse.
- Your data stays put Optional federated learning lets each site improve its models without its data ever leaving the site.
- Sovereign by design Runs entirely on your own infrastructure, with nothing leaving your control.
Why it matters
Critical AI infrastructure has to be robust, resilient and always available, even when part of it fails. Dune is built for that.