Summary
RackNerd is launching United States data center AI VPS with an annual subscription price of $49.95.$21.99To the$119.99Support for the deployment of self-hosted tools such as Hermes Agent, n8n, Ollama , OpenClaw and LibreChat in a CTRL panel key. The mobile plan is suitable for placing AI assistant, workflow, knowledge base and Docker management on the same low to medium allocation VPS .
What It Is
RackNerd is no stranger to VPS readers overseas, where the usual form of product was an occasional release of multiple United States data center yearly releases for purposes such as construction sites, cross-border operations and development testing. The new AI VPS . mobile plan still belongs to the annual model United States .The core change lies in: the official inclusion of Hermes Agent, n8n Ollama , OpenClaw, LibreChat and other AI with automation applications into CTRL panel deployment scope.
For the reader, this type of mobile plan combines a combination of buying an United States and building self-hosted AI workflows into the same purchase. It is not simply oriented towards training or high concurrent reasoning large model server , original unpublished GPU , memory types, number of IPv4 , open runs, payment and refund details; rather it's more suitable for use with CPU built-in lightweight AI assistants, automation platform knowledge base, Docker management and entry to middle age in conventional build site environments.
How to Choose
Plan Specs
| Programs | CPU and | Memories | and SSD | traffic and | bandwidth and | The Price |
|---|---|---|---|---|---|---|
| 1GB of plans | 1 core | 1 GB of memory | 20GB of the SSD | 3 TB | 1 Gbit/s | $21.99/ year |
| The 2GB plan | Nucleus two | 2GB of storage | 35GB and SSD | 5 TB | 1 Gbit/s | $35.99/ year |
| 4GB plan | 3 to the core | 4GB of storage | This is a 60GB SSD . | 7 TB | 1 Gbit/s | $59.99/ year |
| The 6GB plan | 6th core | 6GB of storage | 100GB and SSD | 12 TB | 1 Gbit/s | $89.99/ year |
| 8GB plan | 7th core | 8GB of storage | 150GB SSD or more | 20 TB | 1 Gbit/s | $119.99/ year |
data centers and
| optional data center | State of affairs |
|---|---|
| The New York | List of titles |
| This is Dallas. | List of titles |
| I'm going to Chicago. | List of titles |
| San Jose | List of titles |
| I'm going to Atlanta. | List of titles |
| Los Angeles, United States and Canada | List of titles |
APP Minimum Configuration
| Other uses | Used | Recommended minimum configuration | Summary of the report |
|---|---|---|---|
| OpenClaw | AI assistant | 4+ vCPU / 8+ GB RAM | Self-hosted AI assistant for automating tasks, workflows, messages and connectivity services |
| The Hermes agent | The AI agent | 2+ vCPU / 4+ GB with RAM | Self-hosted AI agent with web management interface, support for chat, tools, message gateways and automation |
| n8n | Workflow | 2+ vCPU / 2+ GB with RAM | Self-hosted automated platform, connectable applications, API , databases and various services |
| Anything LLM | Knowledge base | 2+ vCPU / 4+ GB with RAM | Self-hosted AI workspaces supporting document dialogue, agents and private knowledge bases |
| Ollama and | Located at large model | 2+ vCPU / 4+ GB with RAM | For deploying and running the open source language model, a chat interface is provided with an Open WebUI in the official description |
| Free Chat | AI chat | 2+ vCPU / 4+ GB with RAM | Self-hosted multi AI server chat interface, support for accounts and chat logs |
| Portainer | Docker management and administration | 1+ vCPU / 2+ GB RAM | Web visualization Docker management tool for managing containers, images and stacks |
| Coolify | Application deployment | 2+ vCPU / 2+ GB with RAM | Self-hosted PaaS , deployable applications, databases and services supporting automated SSL and Git workflows |
| The Webmin | Linux management and administration | 1+ vCPU / 1+ GB RAM | Web Linux system management tool for managing users, services, software packages and system settings |
| The CyberPanel | Admin panel | 1+ vCPU / 1+ GB RAM | Web Hosting Control Panel, which can be used to manage websites, email, DNS , databases and so on. |
| This is CloudPanel | Admin panel | 1+ vCPU / 2+ GB RAM | A modern web hosting panel that supports PHP , Node.js , Python , static sites, NGINX, SSL and MySQL |
Fit Checklist
| Your Needs | Recommended initial configuration | Explained |
|---|---|---|
| Single site, simple admin panel and lightweight Docker | 1GB or 2GB plan | Suitable for Webmin, CyberPanel, CloudPanel and Portainer but requires control over concurrency and containers |
| Automation platforms such as n8n, Coolify | Start of the 2GB program | The official recommendation for n8n is 2+ vCPU / 2+ GB RAM . |
| This is the first time I've been able to do this. | 4GB program launched | The official recommendation is 2+ vCPU / 4+ GB RAM , with a 4GB option closer to the minimum requirement. |
| OpenClaw | 8GB plan launched | Officially recommended 4+ vCPU / 8+ GB RAM , 1GB, 2GB, 4GB and 6GB options are all lower than the recommended memory |
| Multiple tools running for a long time | 6GB or 8GB plan | Multiple applications will occupy CPU memory and disk I/O simultaneously, with low-allocation schemes more prone to resource strain |
Notes and notes
- promo code with activity timeIf the mobile plan promotion is more than 30 days old, the event may have ended. The price, stock and payment page should be considered as official.
- Buy entry: The box in the original form indicates only mobile plan entries, no specific purchase link has been disclosed.
- Payments and refundsPayment method, refund policy, renewal price and post-expiration prices not disclosed.
- Open run scoresModel CPU , disk IOPS, network latency , application of actual throughput or runpoint sources and no performance commitments.
- How AI applications are deployed: When you choose to install an AI application, it needs to be deployed separately through the CTRL panel environment.
- bandwidth caliperTable marked 1Gbps, but bandwidth does not equal guaranteed continuous throughput. Actual speed is affected by data center , network congestion, application load and quality of carrier .
- traffic caliper: The original lists 3TB, 5TB, 7TB, 12TB and 20TB; it is not disclosed whether the calculations are based on monthly, annual or other cycles.
- Regular useThe mobile plan also supports other routine uses of the VPS , such as build stations, development tests, foreign trade business demonstration, automated scripts and self-hosted tools.
- Unsuitable for the sceneNot suitable for users who require a stable high and occuring environment, GPU large model training or high I/O database clusters; unpublished originals of the types GPU , NVMe / SSD , IPv4 quantity and advanced network parameters.
- data center Suggestions for selection: Unpublished data center to region-specific latency data, readers can choose from the regions where business users are located, sources of existing website access and billing habits. Quality should not be judged solely by the name data center .
- Compliance useOnly for legitimate purposes such as site construction, development testing, foreign trade operations and self-hosted applications; does not involve circumvention of regulation, proxy access, node distribution or other uses that do not comply with terms of service .
FAQ and
1. What is it?$21.99Can the 1GB/year plan deploy Ollama ?
It is not recommended to use it as a long-term runtime option. The original Ollama suggested the minimum configuration of 2+ vCPU / 4+ GB RAM , while the 1GB plan only had 1 core and 1GB, below the recommended configuration. Short term tests can be installed but multi-model, multi-user or ongoing tasks are more prone to encountering memory and CPU bottlenecks.
2. Why does OpenClaw recommend the 8GB option?
The minimum recommended configuration for OpenClaw is 4+ vCPU / 8+ GB RAM . Only the 8GB option listed in mobile plan reaches 7 cores and 8GB, while none of the other 1GB, 2GB, 4GB or 6GB options have reached the recommended memory.
3. Can AI VPS be used as an ordinary terminal?
The mobile plan also supports other common uses of the VPS and provides panels or deployment tools such as Webmin, CyberPanel, CloudPanel and Coolify. If used for producing websites, it is still necessary to confirm backups, SSL , firewall , software updates and disk balances by yourself.
4. Does the annual price include a renewal fee?
The original text only lists the yearly price, e.g.$21.99/year$35.99/year$59.99/year$89.99/year$119.99/ year, undisclosed renewal price, expiry price and automatic renewal rules. Refer to the official account page before purchasing
5. Can other cities be substituted for the names in data center ?
The options listed in the original article are data center for New York, Dallas, Chicago, San Jose, Atlanta and Los Angeles. Whether they can be substituted or whether all mobile plan s support the same data center , needs to be shown on the official order page as a matter of fact.
Related links
- RackNerd United States compared to
- Guide to selecting self-hosted AI tools
- traffic , bandwidth and CPU entrance
- Annual payment VPS Checklist of terms for refunds
- United States data center latency with a selection of business areas
What time is it?
This article was last updated on 06/22/2016.
This information is current as of 2026-06-22; prices, inventory, data center availability, refund policy and application list are official.
Image by Alt
- alt: flat vector server racks, AI automation icons, and a map of US data centers
- alt: isometric 3D VPS chip, control panel blocks, and one-click deployment arrows
- alt: card, SIM-free device, CPU cores, and storage blocks
- alt: data center corridor with blue server lights and network lines
Reader comments (0)