Pay for AI. Pay for servers.
One invoice
Hydron is in Closed Beta. Pricing shown here won’t apply until 30 days after your beta period begins. You’ll build and deploy for free while you’re getting started.
Free creditsActive
Try the platform on us
Cancel anytime
No long-term contract
Take it anywhere
Your config, your control
Here’s what’s on your bill
AI credits, bare metal server costs, and optional top-ups broken down below.
Your monthly bill
One invoice · pay only for what you useSubscription
Required$20 / mo
Includes $20 of credits
Activates your account and refills your credit balance every billing cycle.
- $20 of credits refilled every period
- Conversational deployment via Hydron chat
- Automatic Docker, SSL, domain & CI/CD setup
- Unlimited projects, chats, and deployments
- Cancel any time, no lock-in
Dedicated servers
When you provisionFlat fee
per server, monthly
Real production-grade servers that Hydron provisions, deploys to, and manages for your apps.
- One flat monthly fee per server in your inventory
- Pricing shown up front when you pick a tier
- Full root SSH access to a dedicated server
- Automatic build, deploy, SSL & domain wiring
- Add or release servers any time
- Take your configs with you
Credit top-ups
Optional$5+
up to $500 per top-up
Need more AI capacity than your monthly allowance? Add credits to your balance any time.
- Available to active subscribers
- Charged exactly as credits are spent
- Optional auto top-up with monthly cap
- Topped-up credits never expire
- Live transaction history & cost breakdowns
- Credits stay even if you cancel
See how far your credits go
Pick a budget and a model. We'll show how many real Hydron actions that typically covers, based on measured averages from actual platform usage.
$5–$500
$USD
Model
Typical capacity with Claude Sonnet 4.6
$20 in credits
Full codebase analysis
≈ $5.73 per action · ≈ 1750K in / 32.0K out tok
~3
actions
Provisioning plan
≈ $1.71 per action · ≈ 375K in / 39.0K out tok
~11
actions
Infrastructure plan
≈ 25¢ per action · ≈ 75K in / 1.9K out tok
~80
actions
Deployment
≈ 5¢ per action · ≈ 16K in / 0.4K out tok
~400
actions
Token counts are median values measured from real production-scale Hydron actions (dev/test runs excluded). Your real cost scales with repo size and project complexity.
Estimated server costs
Each server you add has a flat monthly fee based on its resource tier, and you see the price before you commit. Add or release servers any time; you only pay for what’s in your inventory.
Hobby
99.9% SLAfrom$10/ mo
- Max RAM
- 256 GB
- Public bandwidth
- 300 Mbps
- Private bandwidth
- None
Best for
Hobby, dev/test, personal
Startup
99.9% SLAfrom$30/ mo
- Max RAM
- 512 GB
- Public bandwidth
- 1 Gbps
- Private bandwidth
- 1 Gbps
Best for
Startups, small hosting
Growth
99.9% SLAfrom$70/ mo
- Max RAM
- 1 TB
- Public bandwidth
- 2 Gbps
- Private bandwidth
- 2 Gbps
Best for
Business apps, gaming, backup
Most popular
Business
99.95% SLAfrom$98/ mo
- Max RAM
- 576 GB
- Public bandwidth
- 5 Gbps
- Private bandwidth
- 25 Gbps
Best for
SME production workloads
Scale
99.99% SLAfrom$472/ mo
- Max RAM
- 1.5 TB
- Public bandwidth
- 10 Gbps
- Private bandwidth
- 100 Gbps
Best for
HPC, Kubernetes, large DBs
Enterprise
99.99% SLAfrom$1,082/ mo
- Max RAM
- 3 TB
- Public bandwidth
- 10 Gbps
- Private bandwidth
- 50 Gbps
Best for
Mission-critical, HCI, AI/ML
Starting prices reflect the entry configuration of each tier. Exact pricing is shown up front when you pick a server during provisioning.
Frequently asked questions
An action is one complete AI-driven step Hydron runs for you, like analyzing your codebase, planning infrastructure, generating a provisioning plan, or executing a deployment. Each one is made up of one or more LLM calls, and credits are deducted at the exact token rates of the model you picked.
Yes, you'll need an active subscription to use Hydron. The subscription includes $20 of credits each billing cycle.
The $20 monthly subscription unlocks the platform, the AI workflow that analyzes, plans, provisions, and deploys for you. It includes $20 of AI credits each cycle, so light usage is covered by the subscription itself. Anything beyond that, extra AI usage or dedicated servers, is billed on top, only when you use it.
Anything AI-powered in Hydron: codebase analysis, infrastructure planning, provisioning plans, deployments, and chat replies. The capacity calculator above shows how many of each action a budget typically covers, based on real platform measurements.
Yes, as long as your subscription is active. Top-ups must be between $5 and $500 per transaction, with no daily or monthly limit on how many top-ups you do. Auto top-up can also refill automatically up to a monthly cap you set.
Each dedicated server has a flat monthly fee, shown up front, so you know the cost before you provision. You pay through Hydron in one bill, there's no separate hosting account or provider invoice to manage. Servers are billed monthly while they exist; release one any time to stop its charge.
Yes. You can pick any of the supported Claude models per project; Opus for the heaviest analysis and planning, Sonnet for an everyday default, Haiku for fast iterations. Switching models is instant and only affects future calls.
It depends on the credit type. Top-ups expire after 60 days and stay in your account even if you cancel your subscription. The monthly subscription resets each billing cycle, so unused credits don't carry over into the next month.
Yes, any time, with no penalty. You keep platform access through the end of the current billing period.
Cancelling keeps your platform access through the end of the billing period. Dedicated servers are billed separately and run as long as you keep paying for them, so releasing a server, or letting its billing lapse, takes the apps on it offline at the end of the paid period, since that's where they run. Everything runs on standard Docker, so you can export the setup and redeploy your stack elsewhere without rebuilding it.