JCIL Cloud overview
JCIL Cloud · for organizations

Pay only for what you use.
No subscription games.

Prepay $25 to activate your first key. Every chat message deducts from your balance at the per-model rate. Stop paying when you stop using. Volume discount kicks in at $500/mo.

Start with $25Read the docs

Three models. Pick the one that fits.

Prices shown per typical message (~1K input tokens + 500 output tokens). Real cost varies with message length — billing uses actual token usage.

Fast

jcil-fast
$0.01per message
  • High-volume answers
  • FAQ bots, simple Q&A
  • Fastest responses
  • Biblical guardrails included
Recommended

Standard

jcil-standard
$0.03per message
  • Balanced reasoning
  • Best for most deployments
  • Handles nuance well
  • Recommended default

Deep

jcil-deep
$0.10per message
  • Deepest reasoning
  • Complex theology, counseling
  • Longest context window (1M)
  • When quality > speed

Every key includes

JCIL's biblical base prompt — pro-life, Scripture-faithful, can't be jailbroken
Custom identity prompt (2,000 chars)
Up to 10 knowledge-source URLs, cached hourly
Widget branding (title, colors, logo, welcome message)
Iframe embed + floating-bubble script
Origin allow-list (lock the widget to your domains)
OpenAI-compatible REST API for custom integrations
Server-side model routing and key rotation
Rate limit protection (10/min/visitor)
Streaming responses
Usage tracking + cost dashboard
Up to 3 API keys per account
Volume discount

15% off over $500/mo

Heavy deployments automatically get 15% off all usage once your spend crosses $500 in a calendar month. Applied retroactively to that month — no contract, no commitment.

Enterprise

Custom contract

For $5K+/mo deployments: dedicated Slack, custom rate limits (1500+/min), SLA, SSO, priority support, pgvector RAG for 100+ page knowledge bases. Email support@jcil.ai.

Pilot program — closing soon

First 3 pilots get 3 months free.

We're looking for three early organizations — a church, a Christian university, a ministry or counseling practice — to run JCIL Cloud for 90 days free in exchange for feedback and a case study. Direct-to-founder support. Custom onboarding.

Apply for a pilot slot Have a code? Redeem it

What you'd actually pay

Real-world examples based on typical traffic patterns we see.

Small church (500 member site)

200 visitor messages / month, Standard default

~$6 / month

Christian university admissions page

2,000 prospective-student messages / month, mix of Standard + Deep

~$90 / month

Ministry with 10k / month traffic

5,000 messages / month, mostly Standard

~$150 / month (before volume discount)

Counseling practice intake bot

500 messages / month, Deep for careful reasoning

~$50 / month

Christian K-12 FAQ bot

1,500 messages / month, Fast for quick parent Q&A

~$15 / month

Frequently asked

Is there a monthly minimum?+

No. Prepay $25 to activate your first key — that's all. After that, pay per message with no ongoing commitment. If you stop sending traffic, you stop paying.

What happens if I run out of credits?+

The widget gracefully tells visitors it's temporarily unavailable. You add credits from Settings; service resumes instantly. You also get email alerts at 80% and 100% of your monthly cap.

Can I change models per key?+

Yes. Each key lets you pick which JCIL tiers are allowed (Fast / Standard / Deep checkboxes) and which is the default. Per-request model override is supported via the API.

How much of my data gets sent to model providers?+

Only the message content + your system prompt (including your knowledge-source content when relevant). We do not send user PII, IP addresses, or your API keys. API providers do not train on JCIL API traffic.

Can the assistant be jailbroken into anti-Christian content?+

No. Your custom prompt is wrapped INSIDE the JCIL base identity. You can shape tone and persona; you cannot override pro-life, Scripture-faithful, or anti-blasphemy guardrails. If a visitor tries, the assistant politely declines and stays in character.

Do I own my conversations?+

Yes. You can enable conversation logging per key (not on by default for privacy). Logs are yours to export. Visitors are anonymous unless you choose to collect identifiers.

What's the latency like?+

Fast: ~1s to first token. Standard: ~1.5s. Deep: ~2-3s. Streaming makes the response feel instant. Knowledge-source cache warms up after the first request of the hour.

Ready in ten minutes.

Create a key, customize the widget, paste the iframe. First 3 pilots get 3 months free.

Open the console Read the docs