Guide
Track AI API token usage
Prepaid billing only helps if you can see the debit. Token Supply records usage per request, rolls it up in the dashboard, and stops new calls when the wallet hits zero.
Key, run, inspect
Create a plan key
Bind the key to the plan that contains your model.
Point the client
Set the base URL and key in Codex, Claude Code, OpenCode, or an SDK.
Send a short request
One prompt is enough to land a usage row.
Open Usage
Check Today, 24h, 7d, or 30d. Wallet balance sits in the sidebar.
What you can see
The dashboard shows usage and wallet activity across your API keys. Switch Today, 24h, 7d, and 30d to see whether a coding agent is a daily habit or a burst.
Each key is bound to a plan. If several people or agents share an account, give them separate keys so Usage can tell them apart.
Prepaid wallet
Accounts start at $0. Add credit before the first call. Listed usage debits the wallet as requests complete.
At zero balance the API returns 402 Payment Required. That is the cap. Add credit when you want the agent to continue.
From a client to a row
After you change a base URL, send a short prompt and confirm that Usage moved. If nothing appears, check the key, the model name, and the client guide for that app.
Codex · Claude Code · OpenCode · OpenAI SDK
Details
- Where is the usage dashboard?
- Sign in and open Usage. Logged-out visitors can still set up a client from the guides.
- What does 402 mean?
- The wallet is empty. Add credit, then retry.
- Can I track more than one agent?
- Yes. Create a key per agent or person. Usage is recorded per key.
Related
- AI API token cost calculator — Compare listed Token Supply rates with public list prices.
- Reduce coding-agent API costs — Keep your client and lower token spend on the same workload.
- Codex custom endpoint — config.toml, auth.json, and the Responses API.
- Claude Code custom endpoint — ANTHROPIC_BASE_URL, auth token, and settings.json.
- Cursor OpenAI-compatible API — Override the OpenAI base URL and paste a plan key.