Ecosystem
Xinference 4.0.1 adds an RCE defense you must switch on
Community reports 4 sources
X
Xinference
Project Oct 16, 2026
A self-hosted model-serving platform with local and distributed deployments and OpenAI-compatible APIs.
Visit primary link
2 sources attached
What happened lately
Xinference timeline
Xinference serves language, image, audio and other models through a common application layer. Its English README documents a Python installation with pip install xinference, local startup with xinference-local, containers and distributed workers. It is serving infrastructure, rather than a model or a coding-agent plugin. Community self-hosting is distinct from the project’s commercial enterprise, cloud and hosted API offerings.
Version 4.0.1 requires Xoscar 0.12.0 or newer and integrates optional actor-message authentication. Administrators must set the same strong XOSCAR_AUTH_KEY on every actor peer and coordinate a restart; upgrading alone leaves the legacy unauthenticated transport enabled. HTTP API authentication does not protect actor ports.