> ## Documentation Index
> Fetch the complete documentation index at: https://platform.atlan.com/llms.txt
> Use this file to discover all available pages before exploring further.

> ## Agent Instructions
> To act on Atlan objects, use the Atlan MCP server at https://api.atlan.com/mcp or the atlanai CLI; `atlanai --map json` prints its command map. Run a read-only identity check before any write.
> The docs MCP server at /mcp searches these docs only. It cannot read or change Atlan objects.
> SDK packages: Python `atlanai` (PyPI) and TypeScript `@atlanai/sdk` (npm). Show Python first, then TypeScript.

# FinOps

> Understand how Agent Registry prices each agent run and where each rate comes from.

# FinOps

Agent Registry calculates a cost for every recorded run at the rates your
organization pays. Use it to compare agents on the same basis and to tell a
contract price from an estimate.

## How a run is priced

Agent Registry prices each run as its trace arrives. It reads the token counts
on the trace and multiplies each by the matching rate for the model.

```text theme={null}
cost = input tokens        x input rate
     + output tokens       x output rate
     + cache read tokens   x cache read rate
     + cache write tokens  x cache write rate
```

Token counts come from the model provider's response. Rates come from
**Models & rates**. If Agent Registry cannot price a run, the cost shown is the
one reported with the trace.

<Note>
  The cost is fixed when the run is recorded. Changing a rate later does not
  change runs that are already priced.
</Note>

## Where the rate comes from

Rates are set per workspace, per model. For each of the four rates, the first
source that applies wins.

| Price from | Meaning |
| - | - |
| **Negotiated** | An Org admin entered a rate for this model. **Negotiated · partial** means some rates were entered; the rest use the seller discount or list price. |
| **Seller discount** | The public list price minus the discount set for the model's seller. |
| **List price** | The public list price. Used when no rate is entered and no discount above 0% is set. |
| **Unpriced** | No public list price exists for this model. Enter a rate to price its runs. |

**List price unavailable** means Agent Registry could not look up the public
price at that moment. It does not mean the model has no price.

Only an **Org admin** can change rates. Every member can view them in
**Models & rates**.

## Read a cost figure carefully

* Check how a model is priced in **Models & rates** before comparing costs.
  A **List price** cost is an estimate; **Negotiated** and **Seller discount**
  costs reflect your contract.
* Cache tokens are priced differently from plain input tokens. A run that
  reads a large cache can show a low cost and a high token count.
* Cost per run is an average for the selected period. Compare it with run
  volume before treating a change as a trend.
* A value shown as unavailable means the report could not read its source. It
  is not a zero.

## Availability

| Capability | Availability |
| - | - |
| Price runs at your rates | Available |
| Match totals to the provider invoice | Roadmap |
| Reprice past runs | Roadmap |
| Budgets and controls | Roadmap |
| Optimization | Roadmap |
| Cost per finished piece of work | Roadmap |

## Learn more

<CardGroup cols={2}>
  <Card title="Models & rates" icon="coins" href="/admin/finops/models-and-rates">
    Connect a LiteLLM proxy and set the rates your organization pays.
  </Card>

  <Card title="Agents usage" icon="chart-line" href="/objects/reporting">
    Compare spend and cost per run across agents.
  </Card>

  <Card title="Agents" icon="bot" href="/objects/agents">
    Inspect an agent's profile, sessions, usage, and recent runs.
  </Card>

  <Card title="Traces" icon="wave-pulse" href="/objects/traces">
    Verify that execution evidence reached the intended scope.
  </Card>
</CardGroup>