# aion ## Docs - [Notifications & Invitations FAQs](https://docs.aion.xyz/FAQs/notification-faqs.md): Common questions about invitations, team access, and in-app notifications. - [Pricing FAQs](https://docs.aion.xyz/FAQs/pricing-faqs.md): Answers to common questions about wallet billing, usage calculation, auto top-ups, invoices, and payments. - [Support FAQs](https://docs.aion.xyz/FAQs/support-faqs.md): Troubleshooting help for instance creation, connectivity, stuck states, availability, GPU options, and SSH key management. - [Delete GPU cluster](https://docs.aion.xyz/api-reference/api/v1/accounts/projects/gpu-clusters/externalId/delete.md) - [Get GPU cluster](https://docs.aion.xyz/api-reference/api/v1/accounts/projects/gpu-clusters/externalId/get.md) - [List GPU clusters](https://docs.aion.xyz/api-reference/api/v1/accounts/projects/gpu-clusters/get.md) - [Create GPU cluster](https://docs.aion.xyz/api-reference/api/v1/accounts/projects/gpu-clusters/post.md) - [Uninstall app from cluster](https://docs.aion.xyz/api-reference/api/v1/accounts/projects/kubernetes/clusterId/apps/appId/delete.md) - [Get installed app details](https://docs.aion.xyz/api-reference/api/v1/accounts/projects/kubernetes/clusterId/apps/appId/get.md) - [Update installed app](https://docs.aion.xyz/api-reference/api/v1/accounts/projects/kubernetes/clusterId/apps/appId/put.md) - [List installed apps on a cluster](https://docs.aion.xyz/api-reference/api/v1/accounts/projects/kubernetes/clusterId/apps/get.md) - [Install app on a cluster](https://docs.aion.xyz/api-reference/api/v1/accounts/projects/kubernetes/clusterId/apps/post.md) - [Delete Kubernetes cluster](https://docs.aion.xyz/api-reference/api/v1/accounts/projects/kubernetes/clusterId/delete.md) - [Get Kubernetes cluster](https://docs.aion.xyz/api-reference/api/v1/accounts/projects/kubernetes/clusterId/get.md) - [List Kubernetes clusters](https://docs.aion.xyz/api-reference/api/v1/accounts/projects/kubernetes/get.md) - [Create Kubernetes cluster](https://docs.aion.xyz/api-reference/api/v1/accounts/projects/kubernetes/post.md) - [List Security Groups](https://docs.aion.xyz/api-reference/api/v1/accounts/projects/security-groups/get.md): Lists all security groups for a project with optional filtering and pagination - [Delete a Security Groups](https://docs.aion.xyz/api-reference/api/v1/accounts/projects/security-groups/id/delete.md): Deletes a security group and all its associated rules. - [Get a Security Groups](https://docs.aion.xyz/api-reference/api/v1/accounts/projects/security-groups/id/get.md): Retrieves details of a specific security group including all its rules - [Update a Security Groups](https://docs.aion.xyz/api-reference/api/v1/accounts/projects/security-groups/id/put.md): Updates a security group's name, description, metadata, and/or rules. - [Create a Security Group](https://docs.aion.xyz/api-reference/api/v1/accounts/projects/security-groups/post.md): Creates a new security group with optional rules for a project - [Revoke Role Assignment from a User](https://docs.aion.xyz/api-reference/api/v1/accounts/roles/assignments/delete.md): Remove a specific role assignment from a user. - [Assign a Role to a User](https://docs.aion.xyz/api-reference/api/v1/accounts/roles/assignments/post.md): Assign a specific role to a user within an account or project - [Update Role Assignment for a User](https://docs.aion.xyz/api-reference/api/v1/accounts/roles/assignments/put.md): Update an existing role assignment for a user. This replaces the current role with a new role. - [Get all roles for an account](https://docs.aion.xyz/api-reference/api/v1/accounts/roles/get.md): Retrieve all roles available within a specific account, including both system roles and custom account-specific roles - [Get a Specific Role](https://docs.aion.xyz/api-reference/api/v1/accounts/roles/types/get.md): Retrieve details of a specific role by its identifier - [Get all Roles assigned to a User](https://docs.aion.xyz/api-reference/api/v1/accounts/roles/users/get.md): Retrieve all roles assigned to a specific user within an account, including account-level and project-level role assignments - [Users in Account](https://docs.aion.xyz/api-reference/api/v1/accounts/users/get.md): Retrieves all users associated with the specified account. Requires authentication and account access verification. - [Invite User to Account](https://docs.aion.xyz/api-reference/api/v1/accounts/users/invite/post.md): Invites a user to join the account by email. - [Revoke API key](https://docs.aion.xyz/api-reference/api/v1/accounts/users/keys/delete.md): Revokes an API key, making it permanently unusable. - [List API keys](https://docs.aion.xyz/api-reference/api/v1/accounts/users/keys/get.md): Lists all API keys for the current account (without sensitive data). - [Create API key](https://docs.aion.xyz/api-reference/api/v1/accounts/users/keys/post.md): Creates a new API key for machine-to-machine authentication. - [Get Profile](https://docs.aion.xyz/api-reference/api/v1/accounts/users/profile/get.md): Fetches a user's profile from the database by their external ID. This endpoint does not call Auth0 and is intended to be used behind a gateway. - [Update Profile](https://docs.aion.xyz/api-reference/api/v1/accounts/users/profile/update.md): Updates a user's profile (currently only name). This endpoint is intended to be used behind a gateway. - [Current User Profile](https://docs.aion.xyz/api-reference/api/v1/auth/me/get.md): Retrieves the authenticated user's profile from Auth0 and local database. - [Validate credential](https://docs.aion.xyz/api-reference/api/v1/auth/validate/get.md): **Port**: 9002 (private) - Internal use only, not exposed publicly. - [Check Authorization for an Action](https://docs.aion.xyz/api-reference/api/v1/authorize/post.md): Evaluate whether a user is authorized to perform a specific action on a resource. The proxy path represents the original request path being authorized. - [Get marketplace app details](https://docs.aion.xyz/api-reference/api/v1/marketplace/kubernetes/apps/appName/get.md) - [List marketplace apps](https://docs.aion.xyz/api-reference/api/v1/marketplace/kubernetes/apps/get.md) - [Update user](https://docs.aion.xyz/api-reference/api/v1/users/put.md): Updates a user's information by their UUID - [Email Verification](https://docs.aion.xyz/api-reference/api/v1/webhooks/auth0/email-verified/post.md): **Port**: 9001 (public) - Called by Auth0 when a user verifies their email. - [User Signup](https://docs.aion.xyz/api-reference/api/v1/webhooks/auth0/signup/post.md): **Port**: 9001 (public) - Called by Auth0 Post-Registration Action. - [Liveness check](https://docs.aion.xyz/api-reference/healthz/live/get.md): Kubernetes liveness probe endpoint. - [Kubernetes Readiness](https://docs.aion.xyz/api-reference/healthz/ready/get.md): Kubernetes readiness probe endpoint. - [Service Version](https://docs.aion.xyz/api-reference/version/get.md): Returns service name, version, and status. - [Privacy and Security](https://docs.aion.xyz/data-security.md) - [Creating Your First Project](https://docs.aion.xyz/getting-started/create-first-project.md): Create a project to organize compute resources, manage access, and track usage in AION. - [Welcome to AION](https://docs.aion.xyz/getting-started/introduction.md) - [Create a GPU Instance and Connect via SSH](https://docs.aion.xyz/getting-started/quickstart.md): Get your GPU instance up and running in minutes. This guide walks you through creating your first AION GPU instance and connecting to it via SSH. - [Create a Kubernetes Cluster on AION](https://docs.aion.xyz/getting-started/quickstart-kubernetes.md): Get a GPU-backed Kubernetes cluster running on AION, retrieve the admin kubeconfig and connect kubectl. - [Setup and Manage Your SSH Keys](https://docs.aion.xyz/getting-started/ssh-management.md): Upload an SSH public key to securely access GPU instances and manage keys from the AION console. - [Checkpointing Training to Amazon S3](https://docs.aion.xyz/guides/checkpoint-s3.md): Learn reliable ways to save and restore training checkpoints to Amazon S3 so you can resume after preemptions, crashes, or maintenance using the Amazon S3 Connector for PyTorch, PyTorch Lightning, s3fs-fuse, or AWS CLI sync. - [GPU Cluster Observability & Diagnostics](https://docs.aion.xyz/guides/gpu-cluster-observability.md): A comprehensive guide to monitoring cluster health, debugging training bottlenecks, and validating hardware stability using burn tests and NCCL benchmarks. - [Multi-GPU LLM Training on a Single H100 Node](https://docs.aion.xyz/guides/training-runbook.md): A guide to train LLMs using all 8 GPUs on an H100 node, utilizing PyTorch Lightning for distributed computing and Amazon S3 for fault-tolerant checkpoint storage and recovery. - [GPU Observability](https://docs.aion.xyz/observability/observability-gpu.md): Monitor GPU and CPU metrics for your clusters and nodes, adjust time ranges and refresh intervals, and maximize panels for a detailed view. - [Gradio App for Stable Diffusion on Ray Serve](https://docs.aion.xyz/playbooks/gradio-playbook.md) - [Serve Stable Diffusion using Ray](https://docs.aion.xyz/playbooks/ray-playbook.md) - [Deploy vLLM on Kubernetes](https://docs.aion.xyz/playbooks/vllm-playbook.md) - [Create and Manage Your API Keys](https://docs.aion.xyz/security/api-keys.md): Generate API keys for programmatic access, copy them once at creation, and manage or delete keys from the AION console. - [Create and Manage Firewalls](https://docs.aion.xyz/security/firewall.md): Create a firewall for your project and add ingress/egress rules to control network access for your instances. - [Identity and Access Management (IAM)](https://docs.aion.xyz/security/iam.md): Manage organization members and control project access by inviting users, adding them to projects, and assigning project admins. ## OpenAPI Specs - [openapi-k8s](https://docs.aion.xyz/api-reference/openapi-k8s.yaml) - [openapi-gpu-compute](https://docs.aion.xyz/api-reference/openapi-gpu-compute.yaml) - [openapi-iam](https://docs.aion.xyz/api-reference/openapi-iam.yaml) - [openapi-firewall](https://docs.aion.xyz/api-reference/openapi-firewall.yaml) - [openapi](https://docs.aion.xyz/api-reference/openapi.yaml)