Know what every request cost

DigitSpace records the cost of every request as it happens: the model and provider that answered, input and output tokens, the price applied, time to first token and total time. Spend adds up by key, member, model, provider and project in your dashboard.

Start free Read the docs

Request req_8f2c41b9

30 Oct 2026, 14:02:11 SGT · key production-api

Asked forauto

Served byclaude-haiku-5-5 on Anthropic

Input1,840 × $0.10 / 1M

Output212 × $0.50 / 1M

Cost$0.000290

First token · total410 ms · 1.9 s

What each request records

Field	Example
Model asked for	auto
Model and provider that answered	claude-haiku-5-5 on Anthropic
Input and output tokens	1,840 in, 212 out (cached and cache-write tokens counted separately)
Cost	$0.000290
Time to first token, total time	410 ms, 1.9 s
Attempts	Each provider tried, with its status and time
Source	The tool that sent it, from the X-Source or X-Title header

The prompt and answer are kept only if your organisation turns on retention.

Telephone switchboard operators with headsets at a long switchboard, plugging cords into the panel. — In 1953 an operator put each call through by hand. DigitSpace routes every request.
Switchboard operators, NIH Clinical Center, 1953. NIH History Office, public domain, via Wikimedia Commons.

Where spend shows up

Activity: Every request, newest first, with its cost and details.
Usage: Spend over time by model, provider, key and project.
Members and teams: What each person and team has spent, against their budget.
The API: GET /v1/key returns the calling key's usage, so a script can check its own spend.

Prices you can check

Each request is priced from the model's published price per million tokens at the moment it ran, and the models page lists those prices. Credits are charged at that price; requests on your own provider keys are free on DigitSpace and billed by the provider.

Questions

A telephone exchange room around 1900, with a long wooden switchboard, operator stools and a desk. — General Telephone Company of the Northwest, exchange room, about 1900, public domain, cropped, via Wikimedia Commons.

Is the cost exact?

It uses the provider's token counts and the published price. If a stream breaks off before the provider reports usage, the request is charged on an estimate, and the log says so.

Can I set limits as well as see spend?

Yes, with budgets per key, member and team. See budgets and access.

How long are logs kept?

Request metadata is kept for your usage history. The prompt and answer are only stored if you turn retention on.

Live in a minute.

Create a key, change the base URL, send a request. No card needed with your own provider keys.

Start free