Any Nu fabric, across processes.
Bind a fabric on one process. Reach it from another. Same Refs, same calls, over TCP or a Unix socket.
See it.
Pair it with nu.mp or nu.cluster to spin the other process. Proxy hosts the fabric there and wires it back to your tree.
Any fabric, hosted in another process.
nu.proxy lets you host any fabric in another process. Your Nu tree flows the same way as if the fabric were right here.
Wrap it with the proxy provider and you are done. No boilerplate, no client to keep in sync.
What you can do with it.
One fabric, hosted once, reachable from anywhere in your system.
shared service
Host a shared fabric.
Run a fabric as its own always-on service. Anyone else can attach and use it whenever they need to.
independent lifetime
Do not die with the main process.
The service lives on its own. Main app restarts, deploys, crashes — the hosted fabric keeps running with all its state intact.
anywhere
Local box or across a rack.
Same wrap, different address. The hosted service does not care where callers live, they just point at it.
Combines well with.
Proxy needs a way to spin the other process. Pair it with a compute fabric on one side and a state or UI fabric on the other.
nu.mp
Local subprocess fabric.
Spawn a local process, install the proxy server in its init tree, done. Zero-dep, no cluster needed.
nu.cluster
Ray cluster fabric.
Same shape as nu.mp, but the process lives on a Ray worker somewhere in the cluster. Proxy wires it back.
nu.kv
Durable state fabric.
Host one KV store. Every other process reads and writes it as a local Ref. No connection pool to manage.
nu.ui
Live browser widgets.
Serve a UI fabric from one process. Other processes push widget state into the same live browser.
Try Nu.
One command gets you the wheel with every fabric. Then follow the movies tutorial to build a real app in an afternoon.
01 Install
pip install "nustack-py[all]"02 Run the demo
nu demo movies03 Build your app
Browse examplesLike what you see?
The project is young. Star it, join the room, watch what we ship next.