JURNITI°
Login
Pricing · last updated 2026

Managed agent hosting,
priced like infrastructure should be.

One agent, one Firecracker microVM, one flat monthly bill. No metered tokens, no markup on your model spend, no surprise line items. Bring your own OpenRouter / OpenAI / Anthropic key and we get out of the way.

See plansOr jump to Hermes

How much does jurniti cost?

jurniti charges a flat monthly fee per agent VM. Starter is $25/month (1 vCPU, 2 GiB RAM), Pro is $49/month (2 vCPU, 6 GiB RAM), and Max is $99/month (4 vCPU, 12 GiB RAM). Go annual and get 2 months free — $250 / $490 / $990 per year. Every plan ships with a 30-day money-back guarantee, BYOK for every model provider, and a real Firecracker microVM — not a container.

Plans

One VM per agent. Pick your size.

Every tier is the full platform — BYOK, no feature gates. Only the box size changes.

Starter

$21/moinstead of $25/mo at the monthly rate

$250 billed once a year — 2 months free

Solo builders running Pi or Hermes 8B with a local proxy. Just enough to keep your agent's home alive 24/7.

1 vCPU · 2 GiB RAM · 10 GiB SSD · 100 Mbit/s burst

  • Real KVM isolation — your own Firecracker microVM
  • BYOK — your keys never leave the box
  • Public TLS subdomain
  • Persistent home that survives restarts

Pro

Most popular
$41/moinstead of $49/mo at the monthly rate

$490 billed once a year — 2 months free

The default. Headroom for Hermes-3-8B at Q4, OpenClaw with a heavy plugin set, or a Pi instance you actually live in.

2 vCPU · 6 GiB RAM · 50 GiB SSD · 200 Mbit/s burst

  • Everything in Starter, plus
  • Custom subdomain
  • 2× the compute, 5× the disk of Starter

Max

$83/moinstead of $99/mo at the monthly rate

$990 billed once a year — 2 months free

For agents you'd cry about losing. Daily snapshots, room for a 12 GiB working set, and the headroom to keep multiple skills warm.

4 vCPU · 12 GiB RAM · 100 GiB SSD · 500 Mbit/s burst

  • Everything in Pro, plus
  • Daily snapshot
  • 2× the compute, 2× the disk of Pro
30-day money-back guaranteeBYOK — no token markupCancel anytimeLive in ~3 minutes
Templates add-on:public templates are free forever; private slots start at $5/mo (5/10/20/50-slot tiers at $5/$9/$17/$29).How templates work
Usage pricing

Or pay by the hour, prepaid.

Same Firecracker microVM, metered per second. Buy credits, run when you need to, stop paying the moment you stop.

On-Demand

usage · prepaid credits

Never interrupted while running.

$0.05/hour

Metered per second while running · Starter sizing

Bursty and part-time agents. Run for an afternoon, stop, pay for exactly the hours you used.

  • Starts in seconds when warm capacity is available, otherwise queued — you'll be emailed when it starts.
  • Running 24/7 ≈ $36.50/mo — the $25/mo flat plan is cheaper for always-on.
  • Prepaid credits — $10 minimum top-up, card on file, never charged until you top up
  • Billed per second while running (60-second minimum per run) at Starter sizing (1 vCPU · 2 GiB RAM)
  • Stop anytime — a stopped agent burns nothing
Buy credits — from $10

Spot

usage · prepaid credits

Half the rate — can be interrupted.

$0.025/hour

Metered per second while running · Starter sizing

Interruption-tolerant work — batch runs, experiments, agents that checkpoint their own progress.

  • AWS reclaims can lose up to the last ~10 minutes of unsynced work; jurniti-initiated pauses lose nothing.
  • Pauses gracefully and restores automatically when capacity frees.
  • Prepaid credits — $10 minimum top-up, card on file, never charged until you top up
  • Billed per second while running (60-second minimum per run) at Starter sizing (1 vCPU · 2 GiB RAM)
  • Stop anytime — a stopped agent burns nothing
Buy credits — from $10

Estimate your usage bill

Drag to your real usage. We’ll show On-Demand, Spot, and the flat plan side by side — and flag the cheapest for you.

Based on a 730-hour month (a full month of 24/7 use) = 122 h/month of metered runtime.

On-Demand$6.08/mo · $0.05/hr · never interrupted
Spot$3.04/mo · $0.03/hr · interruptible
$25 flat plan$25.00/mo · always-on baseline

Cheapest for you: Spot

On-Demand ($6.08/mo) is cheaper than the $25flat plan at this usage — you’re below the ~500 h/mo crossover.

On-Demand beats the $25 flat plan below ~500 h/mo; above that, the flat plan wins. Spot always undercuts On-Demand.

Estimate only. $10 minimum top-up, card charged only when you top up · billed per second while running, 60-second minimum · a stopped agent burns nothing.

Side-by-side

Compare the three plans.

Same platform, three box sizes.

 StarterProMax
Monthly price$25$49$99
Annual (2 months free)$250/yr$490/yr$990/yr
vCPU124
RAM2 GiB6 GiB12 GiB
SSD10 GiB50 GiB100 GiB
Bandwidth100 Mbit/s200 Mbit/s500 Mbit/s
Custom subdomainAuto-assignedCustomCustom
Daily snapshot——Included
Firecracker microVMYesYesYes
BYOK for every modelYesYesYes
30-day money-backYesYesYes
On every plan

What's included before you spend a cent.

The platform doesn't change between tiers. Only the box does.

Real Firecracker microVM

Every agent gets its own KVM-isolated virtual machine. Hardware boundaries, not namespace tricks. The same primitive AWS uses for Lambda.

BYOK for every model provider

OpenRouter, OpenAI, Anthropic, Nous Portal — paste your key into the VM once and your tokens never leave it. We don't proxy, don't see counts, don't mark up.

Public TLS subdomain

<you>.jurniti.com is auto-provisioned and certificate-managed. Plug it into your agent's webhook, dashboard, or chat bridge.

Persistent /persist volume

Skills, memory, conversation history, plugin state — anything you write under /persist survives restart, resize, and snapshot.

30-day money-back guarantee

Cancel your first plan within 30 days for a full refund, no forms, no support gauntlet. The guarantee covers your first subscription and ends once you expand. After that, cancel anytime — your VM keeps running to the end of the period you've paid for, then simply doesn't renew.

Template + fork

Snapshot any VM into a template (private or public). Fork a public template to spin up a configured agent in one Checkout.

What you won't see on the bill

The costs we deliberately don't charge.

Most managed-agent billing pages are minefields. This one isn't.

Token markup

Your model bill goes to your model provider, not to us. BYOK is non-negotiable; we never insert ourselves into the largest line item on your bill.

Per-request fees

No usage metering. No per-invocation charge. Run your agent into the ground at no incremental cost.

Idle-time charges

The VM is yours 24/7. We don't bill differently when it's busy and we don't auto-suspend it for being quiet.

Egress overages

Bandwidth is generous (100/200/500 Mbit/s burst). We don't meter outgoing bytes.

Snapshot or fork fees

Public templates are unlimited and free forever, and forking one into a new VM is a normal Checkout — no snapshot fee, no fork fee. Only private-template storage is paid, via the count-based Templates plan (5/10/20/50 slots at $5/$9/$17/$29 a month).

Onboarding / setup

Auto-provision runs the moment Stripe confirms the payment. No founder-led demo, no setup tax, no onboarding call required to get root.

FAQ

Questions you should ask before paying anyone.

How much does managed agent hosting cost on jurniti?
Starter is $25/month or $250/year (2 months free). Pro is $49/month or $490/year. Max is $99/month or $990/year. Every plan is a flat fee per VM with no token markup, no metered surprises, and a 30-day money-back guarantee.
Is there a free trial or free tier?
No. Spinning up a real Firecracker microVM costs us roughly $0.30–$1.20 a day depending on plan, so a free trial would mean burning cash on browsers who never convert. Instead, every plan ships with a 30-day money-back guarantee on your first plan: cancel within 30 days for a full refund, no questions asked. The guarantee applies to your first subscription and ends once you expand (add a second VM, upgrade, or buy a Templates plan) — by then you've decided we're worth it. You can always cancel anytime; outside the guarantee window, cancelling stops the next renewal and your VM keeps running until the end of the period you've already paid for.
Can I pay by the hour instead of a flat monthly fee?
Yes. On-Demand is $0.05/hour and is never interrupted while running — it starts in seconds when warm capacity is available, otherwise it's queued and you'll be emailed when it starts. Spot is $0.025/hour and can be interrupted: AWS reclaims can lose up to the last ~10 minutes of unsynced work; jurniti-initiated pauses lose nothing — it pauses gracefully and restores automatically. Both run on prepaid credits ($10 minimum top-up, card charged only when you top up) and are metered per second while running, so stopping the agent stops the spend. Running 24/7 ≈ $36.50/mo on On-Demand, so the $25/mo flat plan is cheaper for always-on.
Why a microVM instead of a Docker container?
Docker shares one Linux kernel across every tenant on the box. A kernel bug or container escape in your neighbor's workload can become your problem. Firecracker spins up a real KVM-backed virtual machine per tenant — the same primitive AWS uses to isolate Lambda invocations. Hardware-enforced boundaries, not namespace tricks.
What does BYOK mean in practice?
Bring Your Own Key. Your OpenRouter, OpenAI, Anthropic, or Nous Portal key is pasted once into your VM's environment file and never leaves it. We don't proxy your model traffic, don't see your token counts, and don't mark up your spend. You pay your model provider direct — typically the largest line item on your bill, and one we have no business inserting ourselves into.
Can I run multiple agents on one VM?
No — each agent gets its own microVM and its own subscription. Hermes + OpenClaw + Pi at Starter is 3 × $25/month. We don't bundle harnesses inside one VM because it would break the per-tenant hardware isolation boundary and starve resources at the Starter tier. Each VM is billed independently, so you can mix monthly and annual across agents.
Where do my VMs run?
us-west-2 today, on AWS EC2 nested-virtualization hosts — one density-sliced host per region. Additional regions land when committed regional MRR clears the cost of a second box; we rent capacity ahead of demand only when the math works, so we never have to raise prices because of idle infrastructure.
What happens to my data if I cancel?
Your persist volume (the /dev/vdb mount that holds your skills, memory, and agent state) stays live until the end of your paid period. Before deletion you can fork it into a private template — that creates a snapshot you can restore into a fresh VM later or share publicly. After deletion we keep no backups; the boundary is real.
Can I switch plans?
Yes. Upgrade or downgrade from the dashboard; Stripe pro-rates the difference. Resizing a VM happens at the next restart (a few seconds of downtime). Your persist volume comes with you across the resize.
Do you charge for templates or forks?
Public templates are unlimited and free forever — that's the growth loop. Private templates use a count-based Templates plan: 4 tiers (t5/t10/t20/t50 = 5/10/20/50 private slots) at $5/$9/$17/$29 per month, billed as one SubscriptionItem on your oldest active sub. There is no free private tier. Forking a public template into a new VM is a normal Stripe Checkout — same plan price, no fork fee.

Pick a harness, pick a plan,
be running in three minutes.

Auto-provision runs the moment Stripe confirms your payment. Your VM, your subdomain, and your shell are waiting when the welcome email lands.

HermesOpenClawPi

Or browse community templates to fork a configured agent