Explore Flowstate

The Flowstate platform
Helm

Know what AI costs. See what you get back.

Understand spending, share budgets and connect AI use to the work it supports.

Recent insights All insights

Buy tokens, rent GPUs, or own the rack?

Open-weight models are now good enough to make self-hosting a financial question rather than a capability one. I modelled four options, an office rack, colocation, rented GPUs and per-token APIs, to find where each one actually wins. The answer turns on utilisation, and it isn't the one the internet keeps giving.

Turn a $3m AI bill into $1.9m

Most teams overpay for AI twice without noticing: they run everything through the default flagship model, and they buy the tokens on a metered plan they can't see into. An intelligent proxy fixes both. Here is the maths, a calculator you can drag, and where Flowstate fits.

Start with the work All use cases

Understand AI spending and value

See where the budget goes and what the business gets back.

Share one AI budget across your tools

Fund people, teams and projects across connected services.

Give your team more capacity

Reduce the preparation, checking and follow-up around recurring work.

Connect AI use to delivered work

Follow AI activity into projects, code and business results.
Quick links
Flowstate / People + AI Talk through your priorities
Documentation
Get help
Docs / Integrations

Connect Gemini API and Vertex AI

Connect Gemini API or Vertex AI to see what your Google Cloud billing account spends on them, beside the rest of your AI spend. Both are billed through Google Cloud, so both read your Cloud Billing export in BigQuery. You don’t share a service account key.

You need access to connect AI providers. Ask your Flowstate admin. You also need someone who can manage billing and IAM in Google Cloud.

Choose the connection

Open this in FlowstateCovers
Gemini API (AI Studio)Gemini API spend
Vertex AI (GCP)Vertex AI spend

Connect either or both. The set-up is the same, so the second one is quick.

Before you start

  • Billing Account Costs Manager or Billing Account Administrator on the billing account.
  • BigQuery User on the project that holds the export dataset.
  • The billing export switched on in Billing → Billing export, pointed at a US or EU multi-region dataset.
  • Access for Flowstate set up by your technical team. See For your technical team.

Connect

  1. Go to Settings → Integrations, select Browse catalog and open Gemini API (AI Studio) or Vertex AI (GCP).
  2. Under How to connect, select Use this route.
  3. Enter the GCP project number, Workload Identity Pool id and Pool provider id.
  4. Enter the Reader service account email.
  5. Enter the Billing export table, in the form project.dataset.table.
  6. Optional, for Vertex AI (GCP): to see spend by team, product or anything else you label, enter a label key, such as team, in Split spend by label (optional).
  7. Select Verify and connect.

Check it’s working

The connection is listed under Connected keys with the status Working. See the spend in Agent insights.

What Flowstate brings in

  • Billed cost by product and Google Cloud project, each day, in your billing account’s currency, before any credits.
  • For Vertex AI, if you chose a label, each day’s spend split by that label’s value. Unlabelled spend stays as one line.
  • A total for your billing account. Google doesn’t report cost per person.
  • Flowstate syncs every hour, and brings in whatever history your export holds when you first connect.

If something’s not right

Older spend is missing. The export only holds spend from when it was switched on, or from the start of the previous month for a multi-region dataset.

The connection fails. The pool, the service account’s permissions or the table name is wrong. Select Show what the provider said to see Google’s error.

For your technical team

  1. Ask your Flowstate contact for the issuer and subject for your organisation.
  2. In Google Cloud, go to IAM → Workload Identity Pools. Create a pool and an OpenID Connect provider that trusts that issuer and subject.
  3. Create a service account for Flowstate to read with. Give it BigQuery Data Viewer on the export dataset and BigQuery Job User on the project.
  4. Let the pool use that service account.