Pay for saved tokens.

Adola charges on the input Rose 1 removes before the model call. Start free, validate the receipt trail, then turn on billing when the savings are real.

Start building View API docs

Example monthly input1M tokens

70% removed700K saved tokens

$0.10 / 1M saved tokens x 700K saved = $0.07.

Starter

first 100K saved tokens

Build a workspace, issue a project key, and prove Rose 1 works on real prompts.

1 workspace
Playground runs
Compression receipts
Community support

Start free

Production

$0.10

per 1M saved tokens

Pay only when Rose 1 removes input before your downstream model call.

Project-scoped API keys
Usage and billing rollups
Batch compression
Quota controls

Enable billing

Enterprise

Custom

private deployment and scale

Run Adola with dedicated limits, private networking, and support for regulated teams.

Dedicated throughput
Invoice billing
SLA options
Deployment review

Talk to us

Savings calculator

Estimate the compression hop before you sign up.

Rose 1 pricing follows the receipt: original tokens minus output tokens. Use these examples to decide whether one free playground run is worth your time.

WorkloadInput / monthKeptSaved tokensAdola bill

Support copilotTickets, account notes, policy docs25M30%17.5M$1.75

Agent tracesTool logs, planner state, execution history100M35%65M$6.50

RAG searchRetrieved chunks after reranking250M30%175M$17.50

The downstream model savings depend on provider pricing. Adola only charges on saved input tokens, so the first check is whether your receipts show enough removed context.

Run a free playground receipt

Every plan includes the production path.

The billable event is simple: Adola receives context, Rose 1 returns smaller context, and records the token receipt.

Bearer keys with revocation
Receipts on every request
Provider-neutral text output
Workspace usage limits
Playground testing
Compression ratio controls