Monthly subscription per Team Member. Brain service is separate. Choose any provider. No long-term contracts.
Which Plan Fits You?
Answer three quick questions and we will point you to the right plan.
1
2
3
How many Team Members do you need?
Team Members are your digital workers. Each one handles a specific role.
✓
1 to 3
Solo operator or small team
✓
4 to 5
Growing business with multiple roles
✓
6 or more
Larger organization or many departments
What type of brain service do you prefer?
The brain service powers your Team Members. You can change this later.
✓
Flat-rate subscription
Predictable monthly cost. All usage included.
✓
Pay per use
May cost less for light usage. Variable monthly cost.
✓
Either is fine
Show me both options and let me decide later.
What features matter most?
You can always upgrade your plan as your needs grow.
✓
Just the basics
Core skills, email, and dashboard reporting.
✓
Full skill library and analytics
293 skills, advanced analytics, Casey live chat.
✓
Custom integrations and SLA
For larger operations: unlimited Team Members and sub-users, compliance-grade audit trails, predictive analytics, 4-hour support with SLA, and Cloud Computer Use built in.
Each plan is a monthly subscription per Team Member. Brain service is separate. You choose any provider. Ollama Cloud Pro ($20/mo with $60 included usage credits, USA data residency, 18+ models) is recommended. Per-token providers like OpenAI, Anthropic, Google Gemini, DeepSeek, Mistral, Groq, and others are also fully supported. You set monthly budgets to control spending.
Light
$49.99
per Team Member / month
Plus brain service. From $20/mo total (Ollama Cloud) or varies by provider
Full authentication (email, Google, Microsoft, passkey, SSO, SAML). Multiple brain service providers. Monthly budget controls. Casey quality review. Email outreach. CRM and scheduling. No contracts, cancel anytime.
Cloud Computer Use
Included with the Full Suite and Enterprise plans. Built in for every tier on the local install edition.
Your Team Members can take complete control of a local workstation: reading the screen, moving the mouse, typing, and clicking through any program installed on that computer. A small app called the AEMS Bridge connects the computer to your AEMS team, and every session runs with your approval, your program allowlist, and a tamper-evident log stored on the machine itself.
How It Works
1
Install the Bridge once. Download it from your AEMS settings and sign in. It reaches out to AEMS over an encrypted connection. Your computer never opens a door to the outside.
2
Choose what they may use. Pick program categories: your files, office programs, web tools, business records, or allow everything. The list syncs to the computer on its own.
3
Approve each session. When a Team Member wants control, a window on that machine shows what they will do and how they will confirm it worked. Nothing runs until you approve. Pause or disconnect the Bridge anytime.
4
Watch the record. Every click, keystroke, and result lands in a tamper-evident log on the computer plus a full history in your AEMS dashboard. Screen pictures never leave your machine and are never stored.
What Team Members Can Do on Your Computer
Run any installed program: accounting, scheduling, industry tools, or software with no other connection option
Fill in forms, update records, and enter data at human speed
Read documents and spreadsheets on screen and act on what they see
Use websites and web tools that require a real browser, exactly as a person would
Finish the job and prove it: every task ends with a visible check that the work actually landed
Handle popups, error messages, and interruptions on the screen without getting stuck
Self-host AEMS on your own hardware. Available for Light, Full Suite, and Enterprise tiers. One-year subscription with full data sovereignty. Renew annually to stay current with the latest brain service advancements.
How to Get Started with Local Install
Choose any subscription plan above to create your account. Then visit the Local Install section in your dashboard to purchase a license and download the installer for your platform.
One installer download covers all three styles: you choose the style during setup on your machine. No separate downloads, no rebuilds, no config files to edit.
1
Fully Local (Air-Gapped)
Runs completely offline. No internet needed. Uses local models only. Maximum data sovereignty.
2
Local/Cloud Hybrid
Use local models for simple tasks and cloud models for complex work. AEMS automatically picks the best option for each task. Local for speed and privacy, cloud for heavy reasoning. Your data stays on your machine unless a task specifically needs cloud processing.
3
Locally-Run, Cloud-Connected
Runs on your hardware but uses cloud brain services. Full control of AEMS, cloud power for your Team Members.
Choose Your Brain Service
AEMS works with any language model provider. Use one or mix several. The system automatically routes each task to the best model for the job.
POPULAR CHOICE
Ollama Cloud Pro
Subscription with usage credits: Pro $20/mo includes $60 of monthly usage credits, spent per token across 18+ models including GLM-5.2, DeepSeek V4 Pro, and Nemotron Ultra. Credits refresh monthly; usage beyond credits simply bills at Ollama published per-token rates, or upgrade to Max ($100/mo, $300 credits).
$20/mo
Total for all Team Members
USA Data ResidencyCredits Included18+ Models Included
OpenAI
Per Token
USA
from $0.10/1M input
GPT-5.4, GPT-5.5
Anthropic
Per Token
USA
from $1.00/1M input
Claude Opus 4.8, Sonnet 5
Google Gemini
Per Token
USA
from $0.10/1M input
Gemini 3.1 Pro (2M context)
DeepSeek
Per Token
China
from $0.14/1M input
V4 Flash, V4 Pro
Mistral
Per Token
Europe
from $0.02/1M input
Mistral Large 3, Nemo
Groq
Per Token
USA
from $0.05/1M input
Ultra-fast inference
xAI (Grok)
Per Token
USA
from $0.20/1M input
Grok 4.5
Together
Per Token
USA
varies by model
Open model hosting
NVIDIA NIM
Per Token
USA
from $0.04/1M input
Voice capable, Nemotron
OpenRouter
Per Token
Varies
varies by model
Access to all models
Custom Provider
Per Token
User-defined
varies
Any standard API provider
Local Install also supports local Ollama, vLLM, and llama.cpp servers. Completely free and private.
Choosing a Brain Service
You can use any provider, mix several, or switch at any time. Ollama Cloud Pro ($20/mo with $60 included credits, USA data residency, 18+ models) is recommended for predictable costs. Per-token providers like DeepSeek and Mistral may cost less for light usage. All providers work with all plan tiers.
Expected Monthly Brain Service Cost
Model
Provider
Part-Time
4 hrs/day
Business
8 hrs/day, 5 days
24/7
always on
Estimated monthly brain service cost per Team Member. Part-Time = 4 hours/day, 20 days/month. Business Hours = 8 hours/day, 5 days/week. 24/7 = always active. $20/month Pro subscription (not per Team Member) includes $60 of monthly usage credits, enough for typical small-business teams; sustained 24/7 workloads can exceed credits and then bill per token at Ollama published rates. Local models are free but only available in the Local Install edition. Actual costs vary by workload type and task complexity. Set monthly budgets per Team Member to cap spending.
We refresh provider pricing regularly from public sources, but rates change often. Always confirm current pricing on the provider's own page. Brain service costs are billed by your provider, not by AEMS.
Frequently Asked Questions
What is a Team Member?
A Team Member is a digital worker that handles a specific role in your business: sales, support, scheduling, CRM, marketing, or operations. They work around the clock, learn your business, and are quality-checked by your Team Leader before anything reaches your customers. You pay a monthly subscription per Team Member plus brain service costs.
What is a Brain Service?
A Brain Service is the intelligence that powers your Team Member. It lets them understand tasks, write emails, make decisions, and do real work. AEMS supports any provider: Ollama Cloud (Pro $20/mo includes $60/mo usage credits, spent per token, recommended), OpenAI, Anthropic, Google Gemini, DeepSeek, Mistral, Groq, xAI, Together, NVIDIA, OpenRouter, or any custom API provider. You can connect multiple providers and the system automatically picks the best one for each task.
When am I charged?
You are only charged when you add a Team Member or purchase a local install Service Plan. There is no charge for creating an account or browsing the dashboard. Your subscription activates when you Hire your first Team Member.
What about brain service costs?
Brain service costs depend on your provider. With Ollama Cloud Pro (recommended), your $20/mo subscription includes $60 of monthly usage credits, spent per token across 18+ models. Enough for typical small-business teams. If you exceed the credits, usage simply continues at Ollama published per-token rates, or upgrade to Max ($100/mo, $300 credits). Per-token providers charge based on usage: DeepSeek V4 Flash costs about $10/mo per Team Member, OpenAI GPT-5.4-mini about $15-25/mo, Anthropic Claude Sonnet about $30-50/mo. You set monthly budgets per Team Member to control spending.
Can I cancel anytime?
Yes. AEMS is month-to-month with no long-term contracts or minimum terms. Cancel anytime and you will not be charged for the following month.
What is Local Install?
Local Install lets you run the full AEMS software on your own hardware. Available for any subscription tier. One-time payment for a 1-year license with full data sovereignty. Choose any plan above to create your account, then access the Local Install section from your dashboard to purchase and download.
How does the auto-default model selection work?
When a Team Member starts a task, the system analyzes the workload type and automatically selects the best model from your configured providers. It balances quality (70%) against cost (30%). Enterprise clients get quality-weighted selection; startups get cost-weighted selection. You can override defaults per Team Member in Brain Assignments.
Do I need separate Access Codes for each provider?
Yes. Each provider requires its own Access Code. Add them in the Access Codes page. You can connect as many providers as you want and the system will route tasks to the best one automatically.
Can I use multiple providers at once?
Yes. Add Access Codes for as many providers as you want. The system auto-routes each task to the best model across all your configured providers, falling back if a provider is down or over budget.
How does Cloud Computer Use work, and which plan includes it?
With the Full Suite or Enterprise plans, your cloud-based Team Members can take complete control of a local workstation. Reading the screen, moving the mouse, typing, and clicking through any program installed on that computer. A small app called the AEMS Bridge links the machine to your AEMS team: you approve each session from a window on that computer, you choose which kinds of programs they may use, and every action is written to a tamper-evident log stored on the machine. Your computer opens no doors: the Bridge makes an outward-only encrypted connection to AEMS, and screen pictures never leave your machine. On the local install edition, Computer Use is built in for every tier.
What happens if I exceed my budget?
Budget Guard monitors all model calls in real time. When a per-Access Code or per-Team Member monthly limit is reached, the system pauses that Access Code or Team Member and alerts the manager. Approaching-limit notifications fire at 80% of budget.