Pricing

Pay for seats. Agents run on usage you can see.

Every plan includes a monthly pool of agent runs. Past that, each run bills at its model tier’s rate, and every charge links to the run that caused it.

Save 20%

Starter

For one team trying its first agent workflows.

$19

per seat / month

  • Up to 10 seats and 3 agents

  • 1,000 agent runs / month included

  • Review gates on every action

  • 7-day run log retention

  • Email support

Starter

For one team trying its first agent workflows.

$19

per seat / month

  • Up to 10 seats and 3 agents

  • 1,000 agent runs / month included

  • Review gates on every action

  • 7-day run log retention

  • Email support

Most teams start here

Team

For teams running agents in production.

$49

per seat / month

  • Unlimited seats and agents

  • 10,000 agent runs / month included

  • Approval rules by role

  • 90-day run logs with replay

  • Audit log export as CSV or JSON

Most teams start here

Team

For teams running agents in production.

$49

per seat / month

  • Unlimited seats and agents

  • 10,000 agent runs / month included

  • Approval rules by role

  • 90-day run logs with replay

  • Audit log export as CSV or JSON

Enterprise

For regulated teams that need control at scale.

Custom

annual contract

  • Volume pricing on agent runs

  • SSO and SCIM provisioning

  • EU or US data residency

  • Unlimited run log retention

  • Named support engineer

Enterprise

For regulated teams that need control at scale.

Custom

annual contract

  • Volume pricing on agent runs

  • SSO and SCIM provisioning

  • EU or US data residency

  • Unlimited run log retention

  • Named support engineer

Operations teams at these companies pay for Halden by the seat

Brightmoor

juniper freight

Monolake

Pallas Health

Quarrystone

Verdant Ledger

Kettle & Co.

ostrava/labs

Usage

Runs bill by model tier, per million tokens

Your plan’s included runs cover typical volume. Past that, each run bills the tokens it used at the rate of its model tier. Most runs use Steady.

Swipe the table sideways to see every column.

Model

Input / 1M tokens

Output / 1M tokens

Context

Best for

Swift

Fast, low-cost calls

$0.40

$1.60

128k

Routing, classification

Steady

Default

Balanced default for agent runs

$3.00

$12.00

200k

Most agent runs

Deep

Careful multi-step reasoning

$12.00

$48.00

400k

Long, multi-step reasoning

Prices in USD per 1M tokens. Usage bills monthly, only after your plan’s included runs are used. Failed runs are never billed, and every line on the invoice links to its run log.

Compare plans

Everything in each plan, line by line

Seats bill monthly or annually. Included runs reset on the first of each month and don’t roll over.

Swipe the table sideways to compare all three plans.

Starter

$19 per seat / month

Team

$49 per seat / month

Enterprise

Custom pricing

Seats and runs

Seats

Up to 10

Unlimited

Unlimited

Agent runs included per month

1,000

10,000

Custom

Longest single run

1 hour

24 hours

24 hours

Governance

Review gates on every action

Approval rules by role

—

Audit log export

—

SSO and SCIM

—

—

Data residency (EU or US)

—

—

Retention and support

Run log retention

7 days

90 days

Unlimited

Support

Email

Priority email

Named engineer

FAQ

Billing questions, answered plainly

Can’t find yours? Write to billing@halden.ai and a person replies within one business day.

You pay per seat, monthly or annually. Each plan includes a pool of agent runs every month. Usage past that pool bills at the per-token rate of the model tier each run used, on the same invoice.

A run is one agent carrying one task from trigger to finish, such as triaging a ticket or reconciling one invoice batch. Tool calls, retries and review pauses inside that task are all part of the same run.

Agents keep working and extra runs bill at the usage rates above. Admins get an email at 80% and 100% of the pool, and you can set a monthly spend cap per workspace if you’d rather pause than pay overage.

No. A run that ends in an error is not billed and does not count toward your included runs. The run log still records every step, so you can see where it stopped and replay it.

No. Your tickets, documents and run logs are used only to run your agents. They are not used to train models. Enterprise plans can also keep data in EU or US regions.

Yes. Monthly plans end at the close of the current billing month, and annual plans run to the end of their term. Before you go, you can export every run log and audit record.

Price a pilot on your own queue

Send us one week of ticket or invoice volume. We’ll estimate runs, model mix and monthly cost before you commit to a plan.

Try your brand

Create a free website with Framer, the website builder loved by startups, designers and agencies.