Skip to content
Neutree Documentation
Document
User Guide
Release Notes
Version
v1.0
v1.0.1
v1.1.0
v1.1.1
v1.2.0
Search
Ctrl
K
Cancel
GitHub
Select theme
Dark
Light
Auto
Overview
About this document
Document overview
Documentation update history
Introduction
Product introduction
Viewing the platform overview
Getting started
Getting started
Deployment requirements
Quick deployment
Quick start
Deployment
Deploying the Neutree management plane
Deployment modes
Deployment requirements
Deploying with Docker Compose
Deploying with Kubernetes
Deployment and operations
Installing Neutree on Kubernetes through NodePort
Updating the control plane images
Setting up a temporary image registry
Using an insecure image registry
Configuring the startup timeout for Kubernetes endpoints
Installing NVIDIA Container Toolkit offline
Installing nfs-utils offline on AKE bare-metal nodes
NVIDIA GPU Operator operations guide
Adjusting the HAMi scheduling policy through config_patch
Integrating external monitoring components
Integrating an external Grafana instance
Upgrade
Upgrading Neutree
Upgrading the management plane
Upgrading the CLI tool
(Optional) Upgrading the cluster version
Infrastructure
Basic concepts
Managing image registries
Managing clusters
Managing clusters
Managing static node clusters
Managing Kubernetes clusters
Model service
Concepts
Managing model registries
Managing models
Managing model catalogs
Managing engines
Engine practices
Quickly using the latest engine version in a Kubernetes cluster
Model service practices
Calling an endpoint when using a FunASR model catalog
Calling an endpoint when using a MinerU model catalog
Managing endpoints
Model gateway
Concept
Managing external endpoints
Managing model usage
Managing access logs
Access control
Key concepts
Use cases
Managing users
Managing roles
Managing workspaces
Managing workspace policies
Managing API keys
YAML
Importing YAML
Exporting YAML
Tools
CLI tool
Settings
Customizing the platform appearance
Updating the license
FAQ
Hugging Face model download hangs
Using a non-standard SSH port
Configuring OpenShift compatibility
vLLM v0.24.0 reports "NVIDIA driver too old" during startup
HAMi Scheduler not ready after upgrading or updating Accelerator Virtualization
Router version not upgraded after cluster version upgrade
GitHub
Select theme
Dark
Light
Auto
Neutree Documentation
Redirecting to
latest documentation
…