The outcome

Provide a useful model endpoint without granting consumers model-management, registration, log, cache, or administrative rights.

Xinference is a Open-source platform for launching and serving language, embedding, reranking, image, audio, video, and custom models locally or in distributed clusters. Running and managing multiple open or custom model types behind local or cluster APIs with model lifecycle, resource, authentication, permission, audit, and monitoring controls. This guide narrows that broad capability into one repeatable outcome, with checkpoints that keep the source material and your judgment in the loop.

Before you begin

Set the boundary before the tool starts.

Choose one real task, identify who will use the result, and decide what evidence or test will make the result acceptable. Gather only the source material needed for that task. If the work contains confidential, personal, regulated, or client-owned information, confirm that the platform and account are approved before sharing it.

Troiana principle

AI should make the work easier to inspect. If the workflow removes the source, the owner, or the review step, redesign the workflow.

Step by step

A workflow you can repeat.

  1. 01

    Define task, exact model and license, engine, hardware, context, quality, safety, latency, concurrency, storage, network, user permissions, retention, and rollback targets.

  2. 02

    Install and pin Xinference, initialize the administrator before exposure, persist auth keys and database, then create a user and API key limited to model listing and inference.

  3. 03

    Verify and launch a pinned built-in or custom model with bounded devices and replicas, private access, explicit model UID, and no unreviewed remote or custom code.

  4. 04

    Run representative correctness, safety and load tests through the native and intended OpenAI-compatible clients, recording output parity, latency, memory, recovery, errors, and cost.

  5. 05

    Canary consumers, monitor model and permission activity, rotate keys, archive the manifest, and test stop, relaunch, cache cleanup, rollback, and deletion without granting write access.

Working standard

What good use looks like.

  • Separate inference from model-management rights.
  • Persist and back up auth state.
  • Vet all custom model code.

Xinference 3.0 replaced the legacy auth system with database-backed authentication enabled by default. The first unauthenticated setup call wins, so initialize before exposure; disabling authentication makes every endpoint public. Persist the JWT secret, API-key encryption key, and auth database consistently across API processes. Separate model-read, model-write, registration, log, cache, key, and user permissions; isolate supervisor and worker traffic, vet custom model code, protect logs and monitoring, cap actor recovery, and test migration and restore.

Official references

Check the current product documentation.

Features, plan limits, availability, and data controls change. These official pages are the starting points used for this collection.