# Hyperstack

> Hyperstack is a cloud GPU platform by Nexgen Cloud, offering on-demand access to high-performance GPUs for AI, ML, and computational workloads. Deploy GPU/CPU virtual machines and Kubernetes clusters across three regions (CANADA-1, NORWAY-1, US-1), with block storage volumes, S3-compatible object storage, and high-speed networking. The platform provides SDKs for Python, Go, JavaScript, and Terraform, as well as an MCP server for AI tool integration. AI Studio is an end-to-end Gen AI platform within Hyperstack for taking a model from dataset to deployment in one place, including fine-tuning, evaluation, and inference.

- The REST API base URL is `https://infrahub-api.nexgencloud.com/v1`
- The web console is at `https://console.hyperstack.cloud`
- Authentication uses API keys passed via the `api_key` header
- The documentation MCP server is at `https://docs.hyperstack.cloud/mcp` (read-only search over these pages, no authentication)

## Getting Started

- [Home](https://docs.hyperstack.cloud/docs/intro): Documentation landing page with links to all sections
- [Getting Started](https://docs.hyperstack.cloud/docs/getting-started): Deploy your first VM and connect via SSH, full onboarding walkthrough

## GPU Cloud

### Hardware

- [Hardware Index](https://docs.hyperstack.cloud/docs/hardware): GPU and CPU hardware index: flavors, availability, compatibility, and dedicated cluster requests
- [Flavors](https://docs.hyperstack.cloud/docs/hardware/flavors): GPU and CPU configurations, specs, and regional availability
- [Flavor Features and Restrictions](https://docs.hyperstack.cloud/docs/hardware/flavor-labels): Feature matrix for hibernation, snapshots, bootable volumes, hard reboot, and network optimization by flavor configuration
- [Flavors for Parallel Workloads](https://docs.hyperstack.cloud/docs/hardware/parallel-workloads): Zero oversubscription, NUMA-aware scheduling, and CPU pinning for x4 and x8 GPU flavors
- [GPU Compatibility](https://docs.hyperstack.cloud/docs/hardware/gpu-driver-compability): CUDA, driver, and OS compatibility reference
- [GPU Stock Information](https://docs.hyperstack.cloud/docs/hardware/gpu-stock-information): Real-time GPU availability via API and UI
- [Large Root Disk Flavors](https://docs.hyperstack.cloud/docs/hardware/large-root-disk-flavors): API-only flavors with larger persistent root disks
- [Confidential Computing](https://docs.hyperstack.cloud/docs/hardware/confidential-computing): TEE-based security overview (under development)
- [Request Dedicated GPU Cluster](https://docs.hyperstack.cloud/docs/hardware/gpu-cluster-request): Enterprise dedicated GPU cluster requests

### Virtual Machines

- [Virtual Machines Index](https://docs.hyperstack.cloud/docs/virtual-machines): Virtual machines documentation index: images, snapshots, hibernation, spot VMs, and lifecycle management
- [Virtual Machines Overview](https://docs.hyperstack.cloud/docs/virtual-machines/virtual-machine-features): Complete guide to deploying, configuring, connecting to, and managing GPU and CPU virtual machines across their lifecycle
- [Operating System Images](https://docs.hyperstack.cloud/docs/virtual-machines/images): Pre-installed OS images with CUDA and driver variants
- [Custom Images](https://docs.hyperstack.cloud/docs/virtual-machines/custom-images): Create reusable images from VM snapshots
- [Snapshots](https://docs.hyperstack.cloud/docs/virtual-machines/snapshots): VM snapshot creation for backup and versioning
- [Hibernation](https://docs.hyperstack.cloud/docs/virtual-machines/hibernation): Pause VMs to reduce costs while retaining state
- [Spot Virtual Machines](https://docs.hyperstack.cloud/docs/virtual-machines/spot): Low-cost interruptible VMs for non-critical workloads
- [VM Status and State Management](https://docs.hyperstack.cloud/docs/virtual-machines/manage-vm-state): VM state transitions and cost implications
- [Change VM Flavor](https://docs.hyperstack.cloud/docs/virtual-machines/change-vm-flavor): Change a VM's hardware configuration via snapshot and redeploy workflow
- [Cloud-Init Configuration](https://docs.hyperstack.cloud/docs/virtual-machines/initialization-configuration): Automate VM setup at boot with cloud-config scripts
- [Provisioning Profiles](https://docs.hyperstack.cloud/docs/virtual-machines/provisioning-profiles): Save and reuse VM deployment configurations
- [Console Logs](https://docs.hyperstack.cloud/docs/virtual-machines/console-logs): Access VM boot and initialization logs for troubleshooting
- [Performance Metrics](https://docs.hyperstack.cloud/docs/virtual-machines/vm-performance-metrics-and-events-history): Monitor CPU, memory, disk I/O, and network with event history

### Kubernetes

- [Kubernetes Index](https://docs.hyperstack.cloud/docs/kubernetes): On-demand Kubernetes cluster provisioning
- [Kubernetes Getting Started](https://docs.hyperstack.cloud/docs/kubernetes/getting-started): Deploy and connect to your first Kubernetes cluster in the console
- [Clusters via UI](https://docs.hyperstack.cloud/docs/kubernetes/kubernetes-ui): Deploy and manage clusters through the web console
- [Clusters via API](https://docs.hyperstack.cloud/docs/kubernetes/kubernetes-guide): Create, connect to, scale, and delete Kubernetes clusters via the API
- [Upgrade Kubernetes Version](https://docs.hyperstack.cloud/docs/kubernetes/upgrade-kubernetes-version): Migrate workloads to a newer Kubernetes version by deploying a fresh cluster (no in-place upgrades)
- [CSI Driver](https://docs.hyperstack.cloud/docs/kubernetes/csi-driver): Persistent volume provisioning via Helm chart
- [Ingress for K8s](https://docs.hyperstack.cloud/docs/kubernetes/ingress-k8s): Configure ingress with wildcard domains
- [Firewall Rules for K8s](https://docs.hyperstack.cloud/docs/kubernetes/firewall-rules-k8s): Required firewall rules for cluster node communication

### Storage

- [Storage Index](https://docs.hyperstack.cloud/docs/storage): Storage documentation index: volumes, object storage, NVMe, ephemeral, and high-performance file storage
- [Volumes Index](https://docs.hyperstack.cloud/docs/storage/volumes): Volumes documentation index: shared storage, bootable volumes, creation, mounting, and lifecycle
- [Root Disk](https://docs.hyperstack.cloud/docs/storage/local-nvme): Local NVMe storage for OS and system files
- [Ephemeral Storage](https://docs.hyperstack.cloud/docs/storage/ephemeral): Temporary runtime storage with data loss prevention strategies
- [Shared Storage Volumes](https://docs.hyperstack.cloud/docs/storage/volumes/shared-storage-volumes): Network SSD block storage replicated across multiple servers for data durability
- [Bootable Volumes](https://docs.hyperstack.cloud/docs/storage/volumes/bootable-volumes): Pre-configured volumes with OS images for VM deployment
- [Creating a Volume](https://docs.hyperstack.cloud/docs/storage/volumes/creating-a-volume): Step-by-step volume creation
- [Mounting a Volume](https://docs.hyperstack.cloud/docs/storage/volumes/mounting-volume): Mount and format volumes with filesystem setup
- [Detaching a Volume](https://docs.hyperstack.cloud/docs/storage/volumes/detaching-a-volume): Detach volumes from VMs without deleting them, including hibernation and deletion guidance
- [Volume States and Billing](https://docs.hyperstack.cloud/docs/storage/volumes/volume-states): Volume lifecycle states, available actions in each state, and billing implications
- [Reserved Volumes](https://docs.hyperstack.cloud/docs/storage/volumes/reserved-volumes): Causes, prevention, and recovery for the RESERVED volume state
- [Move Volume Between Environments](https://docs.hyperstack.cloud/docs/storage/volumes/update-a-volumes-environment): Reassign volumes across environments in the same region
- [Object Storage](https://docs.hyperstack.cloud/docs/storage/object-storage): S3-compatible object storage for datasets, logs, and backups
- [High-Performance File Storage](https://docs.hyperstack.cloud/docs/storage/high-performance-file): HPC file storage for intensive workloads (in development)

### Networking

- [Networking Index](https://docs.hyperstack.cloud/docs/network): Networking documentation index: SSH, firewalls, public IPs, bastion servers, and network types
- [SSH Access](https://docs.hyperstack.cloud/docs/network/ssh): Create SSH keys and connect to VMs
- [Connecting to Ubuntu VM](https://docs.hyperstack.cloud/docs/network/ubuntu-ssh): SSH guide for Ubuntu with firewall setup
- [SSH for Multiple Users](https://docs.hyperstack.cloud/docs/network/ssh-multiple-users): Add multiple SSH public keys to a key pair for multi-user VM access
- [Firewalls](https://docs.hyperstack.cloud/docs/network/firewalls): Ingress and egress firewall rules with protocol support
- [Public IP Addresses](https://docs.hyperstack.cloud/docs/network/public-ip): Managing public vs private IPs, assignment, and billing
- [Bastion Servers](https://docs.hyperstack.cloud/docs/network/bastion-servers): Secure bastion topology for accessing internal resources
- [Network Types](https://docs.hyperstack.cloud/docs/network/network-types): Entry-level, standard, high-speed Ethernet with SR-IOV, and InfiniBand network types and which GPU models support each
- [High-Speed Networking](https://docs.hyperstack.cloud/docs/network/high-speed-networking): 350 Gbps inter-VM bandwidth with SR-IOV

### Concepts

- [Resource Management Index](https://docs.hyperstack.cloud/docs/resource-management): Resource management documentation index: environments, regions, dashboard, callbacks, and organization
- [Concepts Index](https://docs.hyperstack.cloud/docs/concepts): Platform fundamentals covering environments, regions, dashboard, and callbacks
- [Dashboard](https://docs.hyperstack.cloud/docs/resource-management/dashboard): Resource overview with active VMs, volumes, and hourly costs
- [Environments](https://docs.hyperstack.cloud/docs/resource-management/environments): Regional resource containers for organizing infrastructure
- [Regions](https://docs.hyperstack.cloud/docs/resource-management/regions): CANADA-1, NORWAY-1, and US-1 data center locations and capabilities
- [Callback URLs](https://docs.hyperstack.cloud/docs/resource-management/callbacks): Webhook notifications for VM and volume state changes
- [API Keys](https://docs.hyperstack.cloud/docs/resource-management/api-keys): Create, rotate, and delete the user-scoped credentials that authenticate Hyperstack and AI Studio API requests

## AI Studio

- [AI Studio Index](https://docs.hyperstack.cloud/docs/ai-studio): Full lifecycle platform for AI models: curate datasets, fine-tune foundation models, deploy inference endpoints, and test in the playground
- [Getting Started with AI Studio](https://docs.hyperstack.cloud/docs/ai-studio/getting-started): Decision hub that routes you to the right quickstart based on what you want to do first
- [Chat with a Model](https://docs.hyperstack.cloud/docs/ai-studio/getting-started/run-a-model): Quickstart for chatting with a hosted model in the Text Playground
- [Generate Images](https://docs.hyperstack.cloud/docs/ai-studio/getting-started/generate-images): Quickstart for generating and editing images in the Image Playground
- [Fine-tune a Model](https://docs.hyperstack.cloud/docs/ai-studio/getting-started/fine-tune-a-model): Quickstart for the full fine-tuning workflow: upload data, train, deploy, and test
- [Get Started with the API](https://docs.hyperstack.cloud/docs/ai-studio/getting-started/api): Quickstart for the OpenAI-compatible API: create a key, make a request, and explore the reference
- [Models Index](https://docs.hyperstack.cloud/docs/ai-studio/models): Models documentation index: base models, fine-tuning, deployment, import, and evaluation
- [Base Models](https://docs.hyperstack.cloud/docs/ai-studio/models/models-overview): Available base models (Llama, Mistral, OpenAI gpt-oss) with context length, modality, and access details
- [Hyperstack Hosted Models](https://docs.hyperstack.cloud/docs/ai-studio/models/hyperstack-hosted-models): Open-weight base models hosted on Hyperstack infrastructure for inference and supported fine-tuning, with per-model context length limits
- [Third-Party Hosted Models](https://docs.hyperstack.cloud/docs/ai-studio/models/third-party-hosted-models): Text-to-text, image-to-text, text-to-image, and image-to-image models offered through third-party provider API integrations
- [Fine-Tuning](https://docs.hyperstack.cloud/docs/ai-studio/models/fine-tuning): Custom model training with LoRA parameter configuration
- [Fine-Tuning Guide](https://docs.hyperstack.cloud/docs/ai-studio/models/fine-tuning-guide): Detailed instructions for training custom models
- [Model Rankings](https://docs.hyperstack.cloud/docs/ai-studio/models/model-ranking): Benchmark evaluation scores across standardized tasks
- [Import Models from Hugging Face](https://docs.hyperstack.cloud/docs/ai-studio/models/import-models): Import LoRA adapters from Hugging Face for inference only; imported models do not support fine-tuning or evaluations
- [Model Aliases](https://docs.hyperstack.cloud/docs/ai-studio/models/model-aliases): Create custom aliases that point to a deployed model endpoint
- [Model Deployment](https://docs.hyperstack.cloud/docs/ai-studio/models/deployment): Deploy fine-tuned models for inference
- [Inference & Playground](https://docs.hyperstack.cloud/docs/ai-studio/playground): Interactive environments for testing text and image model inference, with parameter tuning, side-by-side comparison, and API-ready requests
- [Text Playground](https://docs.hyperstack.cloud/docs/ai-studio/playground/text-playground): Chat-style interface for testing text-to-text and image-to-text (vision) models, with image attachments, conversation history, system prompts, parameter tuning, and side-by-side comparison
- [Image Playground](https://docs.hyperstack.cloud/docs/ai-studio/playground/image-playground): Workspace for generating images from text prompts and editing images with vision models, with side-by-side comparison and API-ready requests
- [Knowledge Bases](https://docs.hyperstack.cloud/docs/ai-studio/knowledge-bases): Index your own text and Markdown files so the Text Playground answers from them and cites the passages used
- [AI Studio Billing](https://docs.hyperstack.cloud/docs/ai-studio/billing): Token-based inference pricing and hourly fine-tuning pricing
- [Logs and Datasets Index](https://docs.hyperstack.cloud/docs/ai-studio/logs-and-datasets): Logs and datasets documentation index: training logs, datasets, and data synthesis
- [Logs](https://docs.hyperstack.cloud/docs/ai-studio/logs-and-datasets/logs): Upload and manage training logs in JSONL format
- [Datasets](https://docs.hyperstack.cloud/docs/ai-studio/logs-and-datasets/datasets): Create filtered datasets from logs
- [Data Synthesis](https://docs.hyperstack.cloud/docs/ai-studio/logs-and-datasets/data-synthesis): Generate synthetic training data from existing AI Studio datasets
- [Evaluations Index](https://docs.hyperstack.cloud/docs/ai-studio/models/evaluations): Evaluations documentation index: benchmark evaluations, custom evaluations, and training metrics
- [Custom Evaluations](https://docs.hyperstack.cloud/docs/ai-studio/models/evaluations/custom-evaluations): Define custom evaluation criteria on selected logs
- [Benchmark Evaluations](https://docs.hyperstack.cloud/docs/ai-studio/models/evaluations/benchmark-evaluations): Standardized tests across math, reasoning, and knowledge
- [Training Metrics](https://docs.hyperstack.cloud/docs/ai-studio/models/evaluations/training-metrics): Monitor training and validation loss during fine-tuning
- [Integrations](https://docs.hyperstack.cloud/docs/ai-studio/integrations): Connect AI Studio to developer tools, editors, coding agents, and AI frameworks using the OpenAI-compatible API
- [Claude Code](https://docs.hyperstack.cloud/docs/ai-studio/integrations/claude-code): Route Claude Code requests through a local proxy to any model hosted in AI Studio
- [n8n](https://docs.hyperstack.cloud/docs/ai-studio/integrations/n8n): Connect n8n workflow automation to AI Studio using the built-in OpenAI-compatible AI Agent node
- [OpenCode](https://docs.hyperstack.cloud/docs/ai-studio/integrations/opencode): Use custom or fine-tuned AI Studio models inside OpenCode from the command line or OpenCode Studio
- [LobeChat](https://docs.hyperstack.cloud/docs/ai-studio/integrations/lobechat): Deploy a self-hosted LobeChat UI backed by AI Studio models
- [LiteLLM](https://docs.hyperstack.cloud/docs/ai-studio/integrations/litellm): Route requests to AI Studio models using the LiteLLM SDK or proxy
- [Zed Editor](https://docs.hyperstack.cloud/docs/ai-studio/integrations/zed-editor): Configure Zed Editor to use AI Studio as a custom AI provider for AI-assisted coding
- [RooCode](https://docs.hyperstack.cloud/docs/ai-studio/integrations/roocode): Configure RooCode in VS Code to use AI Studio models via the OpenAI-compatible API
- [Kilo Code](https://docs.hyperstack.cloud/docs/ai-studio/integrations/kilo-code): Connect Kilo Code in VS Code to AI Studio using the OpenAI-compatible API
- [Cursor](https://docs.hyperstack.cloud/docs/ai-studio/integrations/cursor): Connect Cursor to AI Studio using the OpenAI-compatible API endpoint
- [OpenWebUI](https://docs.hyperstack.cloud/docs/ai-studio/integrations/openwebui): Integrate OpenWebUI with AI Studio using the OpenAI-compatible API

## Build

### Tutorials

- [Tutorials Index](https://docs.hyperstack.cloud/docs/tutorials): Step-by-step guides for GPU workloads and platform features
- [Jupyter Notebook](https://docs.hyperstack.cloud/docs/tutorials/jupyter-notebook): Deploy and run Jupyter Notebooks on GPU virtual machines
- [Docker Support](https://docs.hyperstack.cloud/docs/tutorials/docker-support): Run GPU-accelerated Docker containers on a virtual machine, with cloud-init automation
- [Ollama](https://docs.hyperstack.cloud/docs/tutorials/ollama): Run open-source LLMs on a GPU virtual machine with an OpenAI-compatible API
- [Open WebUI](https://docs.hyperstack.cloud/docs/tutorials/open-webui): Add a browser-based chat interface to an Ollama model on a GPU virtual machine
- [vLLM](https://docs.hyperstack.cloud/docs/tutorials/vllm): Serve open-source LLMs behind an OpenAI-compatible API on a single GPU virtual machine
- [Multi-GPU vLLM](https://docs.hyperstack.cloud/docs/tutorials/vllm-multi-gpu): Serve a large model across multiple GPUs on one virtual machine with vLLM tensor parallelism
- [LLM on Kubernetes](https://docs.hyperstack.cloud/docs/tutorials/llm-on-k8s): Serve a large language model across a multi-node Kubernetes cluster with vLLM
- [Fine-Tuning LLMs with Unsloth](https://docs.hyperstack.cloud/docs/tutorials/fine-tuning-unsloth): Fine-tune Qwen3-8B with QLoRA on a GPU virtual machine using Unsloth
- [Ephemeral Drive Mounting](https://docs.hyperstack.cloud/docs/tutorials/ephemeral-drive-mount): Customize ephemeral drive mount points and filesystems via cloud-init

### Hyperstack MCP Servers

- [Hyperstack MCP Servers Overview](https://docs.hyperstack.cloud/docs/libraries/mcp): Choose between the two Hyperstack MCP servers, the read-only Docs MCP Server and the resource-managing API MCP Server

### API MCP Server

- [API MCP Server Index](https://docs.hyperstack.cloud/docs/libraries/mcp-server): Create and manage Hyperstack resources (VMs, clusters, volumes, billing) through natural language; requires a Hyperstack API key
- [Installation](https://docs.hyperstack.cloud/docs/libraries/mcp-server/installation): Complete installation guide for the MCP server
- [Configuration](https://docs.hyperstack.cloud/docs/libraries/mcp-server/configuration): Configure connection settings, authentication, and environment options
- [Quickstart](https://docs.hyperstack.cloud/docs/libraries/mcp-server/mcp-server-guide): Fastest way to get started: end-to-end walkthrough connecting the MCP server to Open WebUI
- [Tools Reference](https://docs.hyperstack.cloud/docs/libraries/mcp-server/mcp-tools-reference): Available MCP tools and API reference
- [Claude Desktop Setup](https://docs.hyperstack.cloud/docs/libraries/mcp-server/claude-desktop-setup): Configure MCP with Claude Desktop
- [Troubleshooting](https://docs.hyperstack.cloud/docs/libraries/mcp-server/troubleshooting): Common issues and solutions

### Docs MCP Server

- [Docs MCP Server Index](https://docs.hyperstack.cloud/docs/libraries/docs-mcp): Read-only MCP server that answers Hyperstack questions from the documentation and returns citation links; hosted at docs.hyperstack.cloud/mcp, no API key required
- [Tools Reference](https://docs.hyperstack.cloud/docs/libraries/docs-mcp/tools-reference): The five read-only tools (search_docs, get_page, list_sections, lookup_error, get_flavor_info) with parameters and example prompts

### SDKs and Libraries

- [Libraries Index](https://docs.hyperstack.cloud/docs/libraries): SDKs and libraries documentation index: Terraform, Python, JavaScript, and Go
- [Terraform Provider](https://docs.hyperstack.cloud/docs/libraries/terraform): Infrastructure-as-code with HashiCorp Terraform (alpha)
- [Python SDK](https://docs.hyperstack.cloud/docs/libraries/python): Programmatic resource management (alpha)
- [JavaScript SDK](https://docs.hyperstack.cloud/docs/libraries/javascript): JavaScript/TypeScript SDK (alpha)
- [Go SDK](https://docs.hyperstack.cloud/docs/libraries/go): Go SDK for cloud applications (alpha)

## Account Management

- [Organization](https://docs.hyperstack.cloud/docs/resource-management/organization): Team member management, roles, and permissions
- [Email Notification Preferences](https://docs.hyperstack.cloud/docs/resource-management/email-notifications): Choose which email notifications you receive, with required and optional categories and a master opt-out switch
- [Two-Factor Authentication](https://docs.hyperstack.cloud/docs/tutorials/two-factor-authentication): Enable and manage 2FA on your Hyperstack account using an authenticator app

### Billing

- [Billing Index](https://docs.hyperstack.cloud/docs/billing): Billing documentation index: policies, pricing, contracts, VM billing states, snapshots, and vouchers
- [Billing Policies](https://docs.hyperstack.cloud/docs/billing/billing-policies): Account types, billing cycles, payment methods, and usage tracking
- [Resource Pricing](https://docs.hyperstack.cloud/docs/billing/pricebook): How GPUs, CPUs, RAM, storage, and public IPs are priced and billed, and where to find current rates
- [Contracts](https://docs.hyperstack.cloud/docs/billing/contracts): Long-term GPU reservation with discounts and expiration policies
- [VM States and Billing](https://docs.hyperstack.cloud/docs/billing/states-and-billing): Cost implications for ACTIVE, SHUTOFF, HIBERNATED, and DELETED states
- [Snapshot Billing](https://docs.hyperstack.cloud/docs/billing/snapshots): Snapshot storage pricing based on Cloud-SSD rates
- [Vouchers](https://docs.hyperstack.cloud/docs/billing/vouchers): Redeem credit vouchers to add balance to your account
- [Contract Billing Behavior](https://docs.hyperstack.cloud/docs/billing/contract-billing-resource-behavior): Resource billing under contract terms

## Help

- [FAQ](https://docs.hyperstack.cloud/docs/faq): Answers to common questions about VMs, billing, GPUs, networking, and storage
- [Shared Responsibility Model](https://docs.hyperstack.cloud/docs/resource-management/shared-responsibility-model): Security responsibilities split between the platform and the operator
- [Support](https://docs.hyperstack.cloud/docs/support): Support ticket portal and service status page
- [Security Advisory: Copy Fail (CVE-2026-31431)](https://docs.hyperstack.cloud/docs/support/cve-2026-31431): Linux kernel privilege-escalation flaw and Hyperstack VM mitigation steps

## API Reference

- [API Introduction](https://docs.hyperstack.cloud/docs/api-reference/introduction): REST API overview, base URL, request format, and conventions
- [Quickstart](https://docs.hyperstack.cloud/docs/api-reference/quickstart): Deploy and connect to your first GPU VM using the API
- [Authentication](https://docs.hyperstack.cloud/docs/api-reference/authentication): Generate, use, and rotate API keys
- [Errors](https://docs.hyperstack.cloud/docs/api-reference/errors): HTTP status codes, response shape, and common error causes
- [Pagination](https://docs.hyperstack.cloud/docs/api-reference/pagination): How to page through list endpoints
- [Rate Limits](https://docs.hyperstack.cloud/docs/api-reference/rate-limits): Rate limits per endpoint and backoff guidance

### API Endpoints

- [Virtual Machine](https://docs.hyperstack.cloud/docs/api-reference/virtual-machine): Create, start, stop, hibernate, and delete VMs
- [VNC URL](https://docs.hyperstack.cloud/docs/api-reference/vnc-url): Request a browser-accessible VNC console URL for a VM
- [Flavor](https://docs.hyperstack.cloud/docs/api-reference/flavor): List GPU and CPU flavors available in each region
- [GPU](https://docs.hyperstack.cloud/docs/api-reference/gpu): List GPU models and regional availability
- [Stock](https://docs.hyperstack.cloud/docs/api-reference/stock): Real-time GPU stock per model and region
- [Image](https://docs.hyperstack.cloud/docs/api-reference/image): List available OS images and manage custom images
- [Snapshots](https://docs.hyperstack.cloud/docs/api-reference/snapshots): Create and manage VM snapshots
- [Template](https://docs.hyperstack.cloud/docs/api-reference/template): Create and manage reusable Marketplace VM templates
- [Deployment](https://docs.hyperstack.cloud/docs/api-reference/deployment): Execute a Marketplace template to provision infrastructure
- [Profile](https://docs.hyperstack.cloud/docs/api-reference/profile): Save and manage reusable VM deployment configurations
- [User Detail Choice](https://docs.hyperstack.cloud/docs/api-reference/user-detail-choice): Retrieve default flavor and image for new VM deployments
- [Kubernetes Clusters](https://docs.hyperstack.cloud/docs/api-reference/clusters): Provision and manage Kubernetes clusters and node groups
- [Volume](https://docs.hyperstack.cloud/docs/api-reference/volume): Create, attach, detach, and delete block storage volumes
- [Volume Attachment](https://docs.hyperstack.cloud/docs/api-reference/volume-attachment): Attach and detach volumes on a VM
- [Buckets](https://docs.hyperstack.cloud/docs/api-reference/buckets): Manage S3-compatible object storage buckets
- [Access Keys](https://docs.hyperstack.cloud/docs/api-reference/access-keys): Manage object storage access key credentials
- [Object Storage Regions](https://docs.hyperstack.cloud/docs/api-reference/regions): List regions where object storage buckets can be deployed
- [Firewalls](https://docs.hyperstack.cloud/docs/api-reference/firewalls): Create firewall rule sets and attach them to VMs
- [Floating IP](https://docs.hyperstack.cloud/docs/api-reference/floating-ip): Attach and detach public IP addresses on VMs
- [Keypair](https://docs.hyperstack.cloud/docs/api-reference/keypair): Manage SSH key pairs for VM access
- [Environment](https://docs.hyperstack.cloud/docs/api-reference/environment): Create and manage environments for grouping resources within a region
- [Region](https://docs.hyperstack.cloud/docs/api-reference/region): List available regions and their supported capabilities
- [Callbacks](https://docs.hyperstack.cloud/docs/api-reference/callbacks): Configure webhook callbacks for VM and volume state changes
- [Dashboard](https://docs.hyperstack.cloud/docs/api-reference/dashboard): Retrieve an aggregated view of active resources and hourly costs
- [Auth](https://docs.hyperstack.cloud/docs/api-reference/auth): Look up the authenticated user, switch active organization, and manage MFA
- [API Key](https://docs.hyperstack.cloud/docs/api-reference/api-key): Create and delete API keys
- [Organization](https://docs.hyperstack.cloud/docs/api-reference/organization): Retrieve and update organization metadata and manage members
- [Invite](https://docs.hyperstack.cloud/docs/api-reference/invite): Send and manage organization invites
- [RBAC Role](https://docs.hyperstack.cloud/docs/api-reference/rbac-role): Create and manage custom roles from policies and permissions
- [Permission](https://docs.hyperstack.cloud/docs/api-reference/permission): List permissions available for building custom roles
- [Policy](https://docs.hyperstack.cloud/docs/api-reference/policy): List policies available for building custom roles
- [Assigning Member Role](https://docs.hyperstack.cloud/docs/api-reference/assigning-member-role): Assign or revoke a role on an organization member
- [User Permission](https://docs.hyperstack.cloud/docs/api-reference/user-permission): List permissions granted to a specific user
- [Consent](https://docs.hyperstack.cloud/docs/api-reference/user-consent): Review and manage consent decisions recorded on your account, such as auto top-up authorization
- [Email Preferences](https://docs.hyperstack.cloud/docs/api-reference/email-preferences): Read and update which email notifications you receive, individually or all at once
- [Billing](https://docs.hyperstack.cloud/docs/api-reference/billing): Retrieve usage records and billing history
- [Pricebook](https://docs.hyperstack.cloud/docs/api-reference/pricebook): Query resource pricing by flavor and region
- [Vouchers](https://docs.hyperstack.cloud/docs/api-reference/vouchers): Redeem credit vouchers
- [Calculate](https://docs.hyperstack.cloud/docs/api-reference/calculate): Return the hourly billing rate for a specific resource
- [Credit](https://docs.hyperstack.cloud/docs/api-reference/credit): Retrieve account credit balance and low-balance threshold
- [Contract](https://docs.hyperstack.cloud/docs/api-reference/customer-contract): Look up long-term GPU reservation contracts and their terms
- [Payment](https://docs.hyperstack.cloud/docs/api-reference/payment): Initiate payments and retrieve receipts
- [Billing Information](https://docs.hyperstack.cloud/docs/api-reference/user): Set and update account billing details for invoicing
- [Alive](https://docs.hyperstack.cloud/docs/api-reference/alive): Liveness check for the billing API
- [Health](https://docs.hyperstack.cloud/docs/api-reference/health): Liveness check for the object storage API

## AI Studio API Reference

- [Base Models](https://docs.hyperstack.cloud/docs/ai-studio-api-reference/base-models): List available base models, pricing, and details
- [Inference](https://docs.hyperstack.cloud/docs/ai-studio-api-reference/inference): OpenAI-compatible chat completions against deployed models
- [Conversations](https://docs.hyperstack.cloud/docs/ai-studio-api-reference/conversations): Create conversations, send messages with image attachments, rename, and retrieve message history
- [Knowledge Bases](https://docs.hyperstack.cloud/docs/ai-studio-api-reference/knowledge-bases): Create knowledge bases, upload and index text and Markdown files, track indexing, and list knowledge bases available for grounding
- [Datasets](https://docs.hyperstack.cloud/docs/ai-studio-api-reference/datasets): Create, list, and delete datasets for fine-tuning
- [Training Data](https://docs.hyperstack.cloud/docs/ai-studio-api-reference/training-data): Upload and manage training logs and data tags
- [Fine-tuning](https://docs.hyperstack.cloud/docs/ai-studio-api-reference/fine-tuning): Create and manage fine-tuning jobs and model imports
- [Models](https://docs.hyperstack.cloud/docs/ai-studio-api-reference/models): Deploy, list, and manage models and aliases
- [Model Evaluations](https://docs.hyperstack.cloud/docs/ai-studio-api-reference/model-evaluations): Run evaluations and retrieve benchmark results
- [Synthetic Data](https://docs.hyperstack.cloud/docs/ai-studio-api-reference/synthetic-data): Generate synthetic training data and track progress
- [System Prompts](https://docs.hyperstack.cloud/docs/ai-studio-api-reference/system-prompts): Create and manage reusable system prompt templates
- [Image Generation](https://docs.hyperstack.cloud/docs/ai-studio-api-reference/image-generation): Generate images from text or from a source image
- [Overview](https://docs.hyperstack.cloud/docs/ai-studio-api-reference/overview): Organization-level summary and resource usage metrics

## Optional

- [Release Notes](https://docs.hyperstack.cloud/docs/release-notes): Platform release history