Platform

Everything for managing your account, credentials, and usage in one place - your organization and projects, API keys, authentication, usage and logs, and billing. Sign in at the ZeroGPU dashboard.

Dashboard

Organizations and projects

API keys

Authentication

Usage and logs

Billing

Dashboard

Your organization, active project, API key, project ID, and a ready-to-run code snippet, all in one place.

Field	What it is
Organization	Your top-level container (your team or company)
Project	The active workload your requests run in
API Key	The credential your requests authenticate with
Project ID	The UUID that scopes requests to the active project
Code snippet	A prefilled request, ready to copy and run

The dashboard prefills the snippet with your project ID. Copy it and send your first request from the Quickstart.

Organizations and projects

An organization is your top-level container - your team or company. A project scopes the workloads you’re running; use separate projects for dev, staging, and production. Each project is isolated: its own API keys, its own usage, its own logs.

Organization	Project	Use
`my-company`	`production`	Live traffic
`my-company`	`staging`	Pre-release testing
`my-company`	`development`	Local experiments

Switch between projects, create new projects, or create new organizations from the navigation dropdown.

Always separate dev from production. A leaked dev key should never be able to touch your production traffic.

API keys

Create, rotate, and revoke the credentials your requests authenticate with. Issue a separate key per environment. Keys follow a one-way lifecycle and use the zgpu- prefix:

Created (shown once) -> Active -> Revoked (permanent)

Action	What happens
Create	A new key is generated and shown once, in full
View	See masked keys and their creation dates
Revoke	The key is disabled immediately; it cannot be undone

Rules of thumb:

One key per environment - dev, staging, and production each get their own.
Rotate on a schedule - create the new key, deploy it, then revoke the old one.
Revoke immediately if exposed - don’t wait.

The full key is only displayed once, at creation. Copy it then and store it securely - if you lose it, revoke it and create a new one.

Authentication

Every request needs the x-api-key header (your API key). The x-project-id header (the project the request runs in) is optional - include it to scope the request to a specific project. A 401 means a bad API key; a 403 means a bad project ID.

Header	Value
`x-api-key`	Your API key from the dashboard
`x-project-id`	Optional - your project UUID from the dashboard
`content-type`	`application/json`

curl https://api.zerogpu.ai/v1/responses \
  -H "content-type: application/json" \
  -H "x-api-key: $ZEROGPU_API_KEY" \
  -H "x-project-id: $ZEROGPU_PROJECT_ID" \
  -d '{ "model": "YOUR_MODEL", "input": "Your input text here..." }'

Status	Cause	Fix
`401`	Missing or invalid API key	Check `x-api-key` - did you copy the full key?
`403`	Invalid project ID or wrong permissions	Verify `x-project-id` matches the dashboard.

API keys belong in server-side code only - never in frontend JavaScript, mobile apps, or version control. To call from a browser or mobile client, proxy the request through a server you control. See Production patterns.

Usage and logs

Usage shows token counts, request volume, and latency trends. Logs show every request with its model, status, latency, and timestamp.

Usage

Real-time trends for the active project - spot cost spikes and latency regressions before they hit your bill.

Logs

The full request history - inspect, filter, and audit individual calls.

Usage metric	What it tells you
Total requests	How many API calls you’ve made
Average response time	How fast the API is responding
Input tokens	Tokens you’re sending to models
Output tokens	Tokens the models are generating

Log field	Example
Model	The model that handled the request
Status	`200`, `400`, `401`, `403`, `420`, `429`, `5xx`
Latency	Time from request to response
Timestamp	When the request was made

Use both together: Usage shows the trend, Logs show the individual requests behind it. See a spike in Usage, then filter Logs to find the requests that caused it.

Billing

Usage-based pricing per million input and output tokens, priced by model - just like frontier APIs. Each model is priced on capability, latency, and compute cost. ZeroGPU is pay as you go: API usage draws down a prepaid credit balance, with no reserved capacity or minimums.

On the Billing page	What it does
Credit balance	Your prepaid balance; API usage is paid from it
Add to credit balance	Add funds with a saved or new payment method
Auto recharge	Optionally top up the balance automatically when it runs low
Payment methods	Add, change, or remove the cards used for credit and auto recharge
Billing history	Review past transactions and their status

Compare per-model rates in the Model Catalog, and estimate spend for your expected volume with the cost calculator.

Get Started

Models

Guides

Platform

Dashboard

Organizations and projects

API keys

Authentication

Usage and logs

Billing

Dashboard

Organizations and projects

API keys

Authentication

Usage and logs

Usage

Logs

Billing

Next steps

Quickstart

Production patterns

Model Catalog

Dashboard

Organizations and projects

API keys

Authentication

Usage and logs

Billing

​Dashboard

​Organizations and projects

​API keys

​Authentication

​Usage and logs

Usage

Logs

​Billing

​Next steps

Quickstart

Production patterns

Model Catalog

Dashboard

Organizations and projects

API keys

Authentication

Usage and logs

Billing

Next steps