Predictable pricing for real agent workloads.

You only pay for real tool execution - not prompts, orchestration, or model reasoning. Pricing scales with real backend execution.

Start free. Upgrade as your agents grow.

Plans based on how your agents grow

As your agents become more capable, usage naturally increases. TengineAI scales with real execution - not complexity.

Free

$0/month

Perfect for prototypes and internal tools.

600 tool calls per month
20 calls per day limit
Community support
Hard monthly limit

Starter

$9/month

Launch your first production agent.

3,000 tool calls per month
100 calls per day limit
Priority support
Hard monthly limit
Most popular for production teams

Growth

$29/month

For real SaaS workloads.

10,000 tool calls per month
No daily limit
Priority support
Overage: $0.004 per call
Configurable monthly cap
Overage enabled

Scale

$99/month

Serious production systems.

100,000 tool calls per month
No daily limit
Dedicated account manager
Overage: $0.003 per call
Configurable monthly cap
Overage enabled

Upgrade instantly. No contracts. Cancel anytime.

How usage is measured

What is a tool call?

A tool call is a real interaction between your AI agent and your backend.

  • Creating or updating records
  • Writing to external APIs
  • Triggering side effects in your systems

Each backend execution counts as one tool call.

What does NOT count as a tool call?

You are not charged for:

  • Prompt generation
  • Model reasoning steps
  • Internal decision logic
  • SDK handshakes
  • Authentication overhead

You only pay when your backend is actually executed.

Overage & Limits

How limits work depends on your plan:

Free & Starter (Hard Limit)

  • Usage pauses when you reach your daily or monthly included calls
  • Upgrade anytime to continue immediately

Growth & Scale (Overage Enabled)

  • Usage continues uninterrupted
  • Overage is billed per tool call at your plan's rate
  • Optional monthly cap to prevent runaway spend

You stay in control - with visibility, alerts, and optional caps.

Example: Publishing a blog post

A typical AI-driven workflow might include:

Fetch trending content→ 1 tool call
Create draft→ 1 tool call
Generate images→ 1 tool call
Publish post→ 1 tool call
Total: 4 tool calls

You are only billed for actual backend interactions.

Real backend executions only. No token or reasoning charges.

Built for safety and predictable growth

Designed to prevent surprise usage spikes.

Daily tool call visibility
Usage alerts
Hard usage caps (optional)
Scoped access by account, project, and user

You stay in control - even as your agents scale.

FAQs

What happens if I hit my limit?

Free and Starter plans pause at their monthly limit. Growth and Scale plans continue with overage pricing (optional caps available).

Can I upgrade or downgrade?

You can upgrade your plan at any time, and upgrades take effect immediately. Downgrades take effect at the start of the next billing cycle, so you keep access to your current plan's limits until then.

Do tool calls roll over?

No. Limits reset monthly.

Do you charge per model token?

No. Model usage is billed directly by your model provider. TengineAI only charges for backend tool execution.

Can I run multiple agents?

Yes. Projects can support multiple agents.

Start free. Upgrade when you're ready.

Give your agents safe, scoped access to real systems - without running your own MCP infrastructure.