Hostinger Hermes Agent Pricing (2026)

Managed plan: what you pay

The managed plan is the “Managed AI Automations” plan, sold at hostinger.com/ai-automation-apps. One subscription covers four apps — Hermes, OpenClaw, n8n, and Paperclip — and you pick one to start with. The price above is the 24-month rate; Hostinger also sells 12-month and month-to-month terms at higher monthly rates. Prices exclude applicable taxes.

Three things to notice before you buy:

• Renewal is higher. The plan renews at the standard renewal rate shown in the card above. Budget for it from day one — promotional pricing is for the first term only.
• The base price is a starting point. The managed plan includes initial AI credits, but heavier usage requires additional credit top-ups. Nobody publishes your usage for you — monitor it in your dashboard for the first month.
• Switching apps does not cost extra. You can move between Hermes, OpenClaw, n8n, and Paperclip under the same subscription. But switching starts the new app fresh — configurations and history do not transfer — so treat each switch as a clean start.

AI credits: the second bill

Hermes runs on two credit systems at Hostinger. This is where the real cost variance lives:

• nexos.ai credits pay for the language models themselves, routed through the Hostinger AI Router (Claude, Gemini, GPT-class models, and others).
• Oxylabs credits pay for web search and data extraction — the agent’s ability to look things up on the web.

The managed plan comes with pre-configured AI credits so you can start immediately, and ready-to-use AI credit add-ons plus web search credits can be selected at checkout. Beyond that, you top up from the Hostinger dashboard as you burn through them.

Hostinger’s support articles publish example consumption rates per one million tokens (shared nexos.ai credit system, documented in their OpenClaw credit guides, which describe the same pool Hermes uses):

• Gemini 3 Flash: 0.5 input / 3 output credits
• GPT-4.1: 2.2 input / 8.8 output credits
• Grok 4: 3 input / 15 output credits
• Claude Sonnet 4.5: 3.3 input / 16.5 output credits
• Claude Opus 4.6: 5.5 input / 27.5 output credits
• Claude Opus 4: 15 input / 75 output credits

Two practical takeaways. First, model choice is a cost lever: a heavy Gemini Flash workload can cost a small fraction of an equivalent Claude Opus workload. Second, the credit balance is shared across Hostinger AI products that use AI credits — if you also run OpenClaw or n8n AI features, they draw from the same pool. The balance can never go below zero; when it runs low, requests are declined and the agent stops generating responses until you top up.

What is still unpublished (honest gaps)

We could not find the following on hostinger.com as of October 9, 2026, so we are not printing them:

• Exact USD prices of additional AI-credit top-up packs for the managed plan. Hostinger says top-ups are available from the dashboard, but no public pack pricing was listed. Check the live cart or your dashboard before committing to heavy usage.
• The exact number of AI credits included in the managed plan. The page says “initial AI credits” and “pre-configured AI credits” — no quantity is given.
• Per-product credit burn for typical tasks. No official “a research brief costs X credits” table exists; the per-token examples above are the closest published data.

If you run a reseller-style comparison site that prints top-up prices, ask where they got them — as of our check, Hostinger had not published them.

Self-hosted cost: the real range

Self-hosting Hermes on a KVM VPS means two bills of your own: the VPS itself plus model usage.

Hostinger’s own cost tutorial puts a typical self-hosted total anywhere from about $6 to over $85 a month (published in their AU tutorial, September 2026). The range is that wide because model usage dominates: a lightly used agent on cloud models sits near the bottom; a heavy Claude Opus workload blows past the top. Hostinger recommends at least 2 vCPU and 8 GB RAM for a stable Docker deployment, which is the tier shown in the card above.

You can also bypass per-token billing on either route: connect your own API keys (pay the model provider directly) or link a ChatGPT Plus/Pro subscription via OAuth, which Hostinger’s managed onboarding supports. Separately, Nous Research itself sells an optional “Nous Portal” subscription for self-hosted users (free pay-as-you-go tier; paid tiers listed in Hostinger’s cost tutorial) — that is Nous’s product, not Hostinger’s.

Renewal math: a realistic first year

Let us walk through what the first term actually costs, in words rather than promises. You pay the promotional 24-month rate upfront — the card above shows the per-month figure and the renewal rate. At renewal, the rate roughly doubles to the standard renewal price, and taxes are added on top of whatever Hostinger quotes.

Then add credits. A casual user — a few tasks a week on efficient models — may stay within the included credits for months. A production user — scheduled jobs, web research, code execution on frontier models — should treat credits as a second subscription and price it from the per-token examples above before committing. The single most common pricing surprise with agent hosting is not the base plan; it is the first credit top-up invoice.

How to keep the bill down

• Default to efficient models for routine work and reserve frontier models for tasks that genuinely need them. The per-token table above shows a 30x+ spread between the cheapest and priciest options.
• Lengthen the heartbeat interval if your agent wakes on a schedule — every wake-up burns credits even when idle.
• Use the OAuth option (ChatGPT Plus/Pro) if you already pay for it — flat-rate billing replaces per-token metering for supported models.
• Watch week one. Autonomous agents can loop on ambiguous tasks. Monitor your first week of usage, set expectations for task complexity, and iterate before scaling up.

FAQ

Ready to deploy? Our Hermes setup guide walks through both routes step by step, AI credits pricing explains the credit system in depth, and the full Hermes guide covers the honest trade-offs.