AppXen AI Builder

Build chatbots, retrieval (RAG) pipelines, and multi-step AI agents with a drag-and-drop visual editor — running entirely in your own AWS account, with models served by Amazon Bedrock. Production-hardened with automatic HTTPS and PostgreSQL. Your flows and data stay in your account — no per-seat fees, no GPU to run.

Coming soon to AWS Marketplace

Want early access? Email support@appxen.ai.

What's included

Pricing

$0.049/hr (≈ $36/mo software) plus Amazon Bedrock usage, billed through AWS — you pay EC2 separately and keep full control of the instance

Recommended instance

t3.small or larger — no GPU needed, inference runs on Amazon Bedrock

Production stack

Visual builder for chatbots, RAG, and agent workflows behind automatic TLS; Amazon Bedrock via an instance role (keyless); PostgreSQL backing store; UFW firewall and unattended security updates on a hardened Ubuntu 24.04 base

Hosted AI builder vs self-hosted: the real cost

Pricing verified June 2026 against flowiseai.com and AWS us-east-1 rates.

Hosted AI-builder plans cap how much you can run and keep your flows on their infrastructure — and you still pay your model provider's API on top. The appliance is flat infrastructure plus metered Amazon Bedrock usage, with no message caps, in your own account.

Option	Price	Caps	Data lives in
Hosted Starter plan	$19/mo	10 agents · 2,000 messages	Vendor
Hosted Pro plan	$49/mo	10,000 messages	Vendor
AppXen AI Builder appliance	~$53/mo + Bedrock usage	No message caps	Your AWS account

Hosted plan prices are on top of your separate LLM API costs. Appliance all-in = $0.049/hr software (≈ $36/mo) + a t3.small (≈ $15/mo) + EBS, plus your Amazon Bedrock token usage.

Cost scenarios

Prototyping · low volume

A hosted Starter plan at $19/mo is a fine, cheap way to experiment. The appliance is worth it once you outgrow the message caps or want everything in your own account.

Production · beyond 10,000 messages/mo

Hosted Pro caps at 10,000 messages and still bills your LLM API separately. The appliance is ~$53/mo flat with no message caps, and Bedrock is billed through your own AWS account.

Own your flows and data

Your flows, stored credentials, and data live in your AWS account and region — exportable and portable, never on a platform you can be priced off of.

When a hosted plan is the better choice

Pick a hosted plan for quick experiments, if you don't have an AWS account, or if you'd rather not run an instance or enable Amazon Bedrock model access. The appliance wins on no message caps, keyless Bedrock billed through your AWS, and full ownership of your flows.

Getting started

Launch with the included CloudFormation template (it attaches a Bedrock IAM role), enable Anthropic Claude model access in your Bedrock console once, wait ~2 min, then browse https://<public-ip>/ and log in as admin with the password from sudo cat /opt/appliance/credentials.txt.

Questions? Email support@appxen.ai.