AI usage and cost
Dates are UTC. A calculated or estimated amount is not a provider invoice. Unavailable means the token count or the historical rate is missing.
Total spending
—
Requests
—
Input tokens
—
Output tokens
—
Cached tokens
—
Average per request
—
Average per 1,000 tokens
—
Active users
—
Most expensive model
—
Highest-spending user
—
Spending
No spending in this range.
Models
| Provider | Model | Requests | Input | Output | Cached | Cost |
|---|---|---|---|---|---|---|
| No model calls in this range. | ||||||
Users
| User | Requests | Models | Input | Output | Spending |
|---|
Calls
| When (UTC) | User | Provider | Model | Operation | Tokens | Calculated | Provider charge | Class | Status |
|---|
0 calls
Pricing
OpenRouter prices come from its models API. OpenAI and Anthropic rates are entered here. A new rate closes the previous open rate; recorded calls keep the rate that applied at the time.
| Provider | Model | Input | Output | From | Through | Source | Verified |
|---|