See every provider's AI spend in one place
You don't run on one provider anymore — and your costs are scattered across a dozen consoles that each count differently and only ever show you the past. Here's how to pull all of it into one running month-to-date total and a forecast, so you see the month forming instead of waiting for the invoice.
A developer ships a feature on a Friday. A loop retries a large prompt more often than anyone intended, a background job keeps running over the weekend, or a traffic spike quietly pushes calls through a pricier model. Nothing breaks. Nothing alerts. The code works. Then the invoice arrives at the end of the month and it's several times what they expected — and by then the money is already spent. There was no single place that would have shown the line bending upward while it was still happening.
That's the trap with AI and compute spend: the damage is silent, it accrues daily, and the one document that finally shows you the total — the invoice — arrives only after you can no longer do anything about it.
Why don’t provider dashboards warn you about a rising bill?
They can’t — every provider’s console is a rear-view mirror. It tells you, accurately, what you have already spent, but none of them tell you where the current month is heading. So the first time a runaway cost becomes a number you can read is after it has finished running.
It gets harder because almost nobody runs on a single provider anymore. A real project leans on Claude for one job and OpenAI for another, generates voices on ElevenLabs, rents a GPU on RunPod, and serves itself through Cloudflare — each with its own login, its own billing cycle, and its own unit of measure. One counts tokens, another characters, another GPU-seconds, another requests and bandwidth. There is no single number anywhere for “what is AI costing me this month,” so even a sharp spike is buried — split across half a dozen dashboards that don’t add themselves up.
What does scattered AI spend actually cost you?
Spreading spend across providers without rolling it up has four concrete costs — even when no single bill looks alarming on its own:
- No total. Answering the most basic question — what am I spending on AI this month — means logging into every console and summing it by hand.
- No forecast. Every dashboard is rear-view; none project ahead, so you learn you overspent only once the invoice lands.
- No comparison. You can’t see which provider is your biggest line item, or that one tripled while another held flat, without building a spreadsheet.
- Apples to oranges. Tokens, characters, GPU-seconds, and gigabytes aren’t the same unit, so even eyeballing the pieces side by side is guesswork.
What is multi-provider AI cost tracking?
Multi-provider AI cost tracking is one view that reads every provider’s usage, converts each provider’s own units into money at its published rates, and adds them into a single running total — kept current day by day, with a forecast of what next month will cost. It’s the answer to all four costs above.
| Provider | Metered in |
|---|---|
| OpenAI | tokens |
| Claude | tokens |
| RunPod | GPU-hours |
| ElevenLabs | characters |
| Cloudflare | requests / bandwidth |
| Gemini | tokens |
The benefit isn’t a notification; it’s a number worth checking. Because the total moves every day and the forecast moves with it, a spike shows up the next time you open the dashboard — days before the invoice, while you can still change something. You catch the runaway Friday loop on Saturday, not on the first of next month.
How does CostCompass track costs across every provider?
CostCompass is a spend-intelligence dashboard built around exactly this. You connect each provider once, and from then on:
- Everything becomes a cost. Each provider’s usage is read from its own API and priced at published per-unit rates, so all those different meters — tokens, characters, GPU-time — become one comparable number.
- One live month-to-date total across every provider you’ve connected — the single number none of the individual consoles can give you.
- A forecast that projects next month from your burn rate so far, so you see the trajectory, not just the history.
- A per-provider and per-model breakdown, so when the total moves you can see exactly which provider — and which model — moved it.
- Your keys are encrypted with your vault password before they reach us, so what we store is ciphertext only you hold the key to. See the security model for exactly how.
Which AI providers can you track in one place?
CostCompass connects to a growing list of providers — see the full set on the providers page, each with a short setup walkthrough. Billing works differently on every one — different keys, different usage APIs, different units and quirks — so the guides below go deeper on how individual providers bill and what it takes to keep a running total for each. Pick the ones you run on:
By provider
- How to Track Claude (Anthropic) API Costs How Anthropic bills Claude by model and token, and how to track usage, forecast next month, and keep your key encrypted in your browser.
- How to Track Google Gemini API Costs How Google bills Gemini across Vertex AI and the Gemini API — and how to track usage, see your exact billed cost, and forecast next month in real time.
- How to Track OpenAI API Costs How OpenAI bills the API by model and token, and how to track org-wide usage, forecast next month, and keep your key encrypted in your browser.
Frequently asked questions
- Which providers can CostCompass track?
- A growing list of the AI and compute services developers build on — model providers like Anthropic's Claude, OpenAI, and Google Gemini, plus voice, GPU, hosting, and analytics services alongside them. The full current list, each with a short connect walkthrough, lives on the providers page. You connect the ones you use and they roll into the same total.
- How can one number cover providers that bill so differently?
- CostCompass reads each provider's own usage data and converts it to a cost at that provider's published rates — tokens for Claude and OpenAI, characters for ElevenLabs, GPU time for RunPod, requests and bandwidth for Cloudflare. Each provider is metered in its own units, but everything lands in your total already converted to money, so a token, a character, and a GPU-second become directly comparable amounts.
- Does it forecast, or just show what I've spent?
- Both. You get a live month-to-date total of what you've spent so far, and a forecast that projects what next month will cost from your burn rate to date. The forecast is the part a provider dashboard never gives you — it's what turns "what have I spent" into "what is next month going to cost."
- Will I know if my spend suddenly jumps?
- Yes — by watching your running total. CostCompass updates day by day, so a sudden jump shows up in your month-to-date figure and pushes the forecast up the next time you open it. You see a spike forming while there are still days left in the month to act on it, instead of finding it on the invoice after it's already spent.
- Where do my provider API keys live?
- Stored on our servers — but encrypted, and only you hold the key. Your keys are encrypted in your browser with your vault password before they're saved, so what we hold is ciphertext, and you're the only one who knows the password that decrypts it (it stays in your browser). When CostCompass fetches your usage, the key is decrypted in your browser and forwarded to the provider through a broker built not to log it, so the plaintext stays out of our database and logs. The security page documents exactly how.
- Why might CostCompass's total differ from a provider's invoice?
- CostCompass computes cost from each provider's published rates at the moment usage is recorded, so the figure is reproducible and doesn't shift when a provider later changes prices. Discounts that don't appear in the raw usage — batch-API rates, automatic prompt caching, promotional credits, committed-use deals — can make the actual invoice a little lower. Treat the CostCompass number as an accurate, slightly conservative running estimate, not a re-billing of the invoice.
- Why use CostCompass instead of checking each provider's own dashboard?
- Each provider's dashboard shows only its own slice, only what you've already spent, and in its own units — so answering "what is AI costing me this month" means logging into a dozen consoles and adding it up by hand, after the fact. CostCompass turns that into one live, forward-looking number — a running month-to-date total, a forecast, and a per-provider, per-model breakdown across everything you connect. You watch the month form instead of reading about it on the invoice.
About the author
Joubert Berger builds CostCompass, a spend-intelligence dashboard that pulls usage from AI and compute providers into one month-to-date total, a forecast, and a per-provider breakdown. This guide reflects how CostCompass reads each provider's own usage API — see the security model for how your keys are handled.
See your whole AI bill in one place
Connect each provider once and watch a single month-to-date total and forecast keep itself current across all of them.