Quick start
Create a key, test a model, and verify the billed request.
Create and copy an API key
If key editing is unavailable, use an existing active key or contact support. Creating a key does not add account balance. The rolling 5-hour, 1-day and 7-day limits are spend limits in USD, not requests per minute.
- Sign in to the console with your account email and password. Complete the authenticator challenge if your account requires it.
- Open API Keys and choose Create API key. Give the key a name you will recognize in Usage.
- Choose an available group if you need a specific model route. Otherwise keep the default assignment.
- Set a USD quota and optional expiry or IP restrictions, then create the key. A blank or zero quota means unlimited key spend, subject to your account limits.
- Use Copy API key to copy the complete value. You can also reveal an existing key. Store it in a server-side environment variable.
Try a model in the playground
An administrator can optionally configure an identity name and knowledge cutoff for a model in Playground. A short instruction then guides its replies when asked; it does not change the underlying model or its training. These instruction tokens count toward normal usage. Unconfigured models receive no added identity instruction.
The playground supports text conversations without attachments. Sent messages and replies are saved to your account and available when you sign in on another device. Unsent drafts stay in this browser tab. A refresh stops a live response and restores saved partial text without resending. Changing the key or model preserves the conversation, and your next message uses the new selection. First text and duration describe your last request, not a provider-wide latency guarantee. A listed model can still reject a request if capacity, balance, quota, or access changes.
- When Playground is enabled, choose it in the website header. If you are signed out, sign in to continue directly to Playground.
- Select an active API key and an enabled model beneath the message input. Saved chats, New chat, and Status & rates are in the left panel. If no models appear, choose another key or ask an administrator to enable models in Playground settings.
- Check Model status for key access and request activity. Reported channel status comes from monitoring and does not guarantee availability for your selected key.
- Open View rates to check final customer prices when available. Normal usage charges still apply when a current quote is unavailable.
- Enter your message and choose Send, or press Enter. Use Shift + Enter for a new line. Stop cancels the response; work already processed may still be billed.
- Open Billed usage to inspect actual charged amounts. New chat starts a separate conversation. Use Your chats to move between saved conversations. Replies continue while you visit another chat or console page. Sent chats remain in your account after sign-out.
Set your key and choose a model
Replace the key placeholder with the complete key you copied. Choose an exact model ID available to that key; the public catalog is a reference and does not establish account access. These variables last for the current terminal session.
Send your first request
The SDK base URL is https://api.kineticrouter.com/v1. A direct HTTP request uses the complete URL https://api.kineticrouter.com/v1/chat/completions. Do not append /v1 twice. These examples make a real, normally billed request.
Verify the answer and billed usage
- Check the returned assistant message. An error response is not a successful answer.
- Open Usage and select a date range that includes the request. Check the displayed timezone.
- Find the request by its model and key name. Review input/output and cache tokens, latency, and actual billed cost.
- If a request is interrupted or times out, check Usage before repeating it. A retry is a separate request and can incur another charge.
Troubleshooting
For support, include the time, model, request ID if supplied, and error status. Never send your API key, session cookie, or password. Error details vary by endpoint; use the returned status and message rather than assuming one universal error shape.
- Authentication failed: copy the complete key again, remove extra whitespace, and check its active status and expiry. Custom keys do not necessarily begin with sk-.
- Model unavailable or access denied: verify the selected key’s group and available model list. Check IP restrictions and provider route availability.
- Quota or balance exhausted: inspect the key’s quota and rolling USD limits, then the account balance or subscription usage. Changing a key quota does not fund the account.
- Request URL not found: use /v1 as the SDK base URL, and /v1/chat/completions for a direct Chat Completions request. Use the protocol supported by your chosen model.
- Rate-limited or temporarily unavailable: wait before trying again and check service status. Avoid repeatedly resending a request that may already have been processed.
- Stream stopped: keep the partial answer and check Usage. Start another request only when you want to retry.
- Playground with an IP-restricted key: the gateway sees the console server’s connection. A key that works from your application may not permit that connection. Keep the restriction unless you intend to change it.