Usage monitoring
Track requests, unlimited token consumption, image request items, and connection pressure across Unlimited Weekly and Unlimited Monthly.
Product / Usage Visibility
Watch unlimited token usage and 4-connection pressure.
Monitor requests, token consumption, images, and 4 concurrent connections alongside rate-limit signals so teams understand the operating boundary of each API access package.
Control plane
Usage Visibility
Token modelUnlimited
Connections4 concurrent
Rate statesWarning and urgent
Threshold warnings
Expose warning and urgent rate states before traffic pressure affects customer-facing workflows.
Limit responses
Return OpenAI-shaped 429 responses when rate pressure needs clients to pause, reduce demand, or route noncritical work later.
Usage breakdowns
Separate traffic by endpoint, model, key, and time period to understand what is driving usage under the package.
Workflow
From setup to launch prep.
Each module is designed to give developers fast integration paths while giving operators the controls they need before traffic scales.
1
Choose the right package
Start with Weekly or Monthly based on your access window, expected usage shape, and support needs.
2
Use delivered API access
Use the delivered API key and OpenAI-compatible base URL while request shape, rate behavior, token usage, and connection pressure are measured in the background.
3
Respond to rate states
Use warning and urgent states to tune model policy, move background work, or reduce demand before pressure reaches users.
4
Review usage patterns
Track which endpoints and models drive usage so teams can improve product behavior over time.
Rate-aware limit response
HTTP/1.1 429 Too Many Requests
x-unlimitedcodex-rate-state: blocked
x-ratelimit-limit-requests: <workspace_request_limit>
x-ratelimit-remaining-requests: 0
{
"error": {
"message": "Rate limit reached for this workspace.",
"type": "rate_limit_error",
"code": "rate_limit_exceeded"
}
}Built for transparent access windows
Usage visibility makes unlimited token consumption concrete. Teams can see what is happening, understand the 4-connection boundary, and adjust traffic before pressure becomes user-visible.
FAQ
Questions teams ask before rollout.
Which usage types are monitored?
Requests, token consumption, image request items, and connection pressure are monitored across Unlimited Weekly and Unlimited Monthly for service health, logs, and rate-limit behavior.
Do you warn teams before access stops?
The API exposes warning and urgent rate states before pressure becomes user-visible. When a hard rate limit is reached, protected endpoints return OpenAI-shaped 429 responses.
Can usage visibility support multiple products?
Yes. Teams can use separate keys and endpoint reporting to understand consumption across apps, services, and environments.
Ready to build?