AimActok BYOK Platform — Usage Tracking & Analytics
Every API call is logged with metadata only. We track what you call and how much, never what you say.
What we track
- Timestamp — when the call was made
- User ID — your account
- Key ID — which of your keys was used
- Vendor — openai, anthropic, etc.
- Model — gpt-4o, claude-haiku-4, etc.
- Token counts — prompt_tokens, completion_tokens, total_tokens
- Status — success or HTTP error code
What we DON’T track
- ❌ Your prompts or messages
- ❌ Your completions or responses
- ❌ Any other payload content
The gateway is a router, not a logger. Your data flows from your server to the model provider, encrypted in transit, never persisted by us.
API access
Get your usage programmatically:
GET /usage
Authorization: Bearer YOUR_JWTResponse:
{
"usage": [
{
"vendor": "openai",
"model": "gpt-4o-mini",
"prompt_tokens": 1234,
"completion_tokens": 567,
"total_tokens": 1801,
"calls": 23,
"success": 22,
"failed": 1
}
],
"total_tokens": 45230,
"total_calls": 312,
"success": 305,
"failed": 7
}Web dashboard
Same data, with a UI. Sign in to see your dashboard.
Export
Coming soon: CSV/JSON export, scheduled email reports, Slack webhooks on usage spikes.
Data retention
- Free tier: 7 days
- Indie: 30 days
- Team: 90 days
- Enterprise: Custom (1 year+)
After retention period, data is automatically purged. You can request manual deletion at any time.
Cost estimation
Since you bring your own keys, you pay model providers directly at their rates. We don’t see your invoices. For cost estimation:
| Model | Input ($/1M tokens) | Output ($/1M tokens) |
|---|---|---|
| gpt-4o | 2.50 | 10.00 |
| gpt-4o-mini | 0.15 | 0.60 |
| claude-sonnet-4-5 | 3.00 | 15.00 |
| claude-haiku-4 | 0.80 | 4.00 |
| gemini-2.5-flash | 0.075 | 0.30 |
| deepseek-v4-pro | 0.14 | 0.28 |
| qwen3-max | 0.40 | 1.20 |
| minimax-m3 | 0.14 | 0.28 |
These are the providers’ official rates. We don’t add a markup.
Last updated: 2026-06-06. Back to overview