What TokenFront does

If your team uses AI platforms (OpenAI, Anthropic, AWS Bedrock, OpenRouter, and others), the bills show up in a handful of different places, measured in different units (tokens, requests), on different billing cycles. TokenFront pulls all of that into one dashboard, so you can answer two questions without logging into four different consoles:

• What are we spending right now, and on what?
• Is anything about that spending unusual, and is there an obvious way to spend less?

The three things it actually does

1. Tracks cost automatically. Once you connect a platform, TokenFront checks in on a schedule and pulls in your usage and cost data. You don't have to export anything or paste numbers into a spreadsheet. See Connecting platforms to get started.
2. Shows you where the money goes. The dashboard gives you the big picture: totals, trends, and which platform costs the most. The Reports page lets you drill all the way down to a single API key on a single day.
3. Watches for you in the background. You don't have to remember to check in. The Intelligence page flags unusual spending patterns, like a spike or a key that suddenly went quiet, and surfaces concrete ways to spend less, like a cheaper model that would do the same job. Optionally, it'll email you instead of waiting for you to check.

What TokenFront doesn't track

TokenFront tracks metered, usage-based spend: the per-token, per-request cost that shows up on your OpenAI/Anthropic/Bedrock/OpenRouter usage bill, pulled straight from each provider's own usage API.

It does not track flat-rate subscription or per-seat plans, things like a ChatGPT Plus/Team/Enterprise seat, a Claude Pro/Team seat, or any other "$X per person per month" product. Providers don't expose that kind of billing information through an API the way they do usage data, so there's currently no way for TokenFront to pull it in automatically. If your team pays for both, TokenFront will accurately reflect the metered API side. Seat and subscription costs won't appear anywhere in the dashboard, Reports, or Intelligence.

Who this is for

Anyone with a stake in what your company spends on AI: an engineering lead who wants to catch a runaway integration before the invoice does, a finance or ops person who wants a monthly number they can trust without asking engineering for it, or a founder who just wants one dashboard instead of five logins.

Where to go next

• New here? Start with Finding your way around.
• Ready to see real data? Go to Connecting platforms.
• Already connected something? Try Setting up your dashboard.
• Want the rest of your team looking at the same dashboard? See Team members.