Metered on the same few units as everything on the network. On your own grams the work is charged and credited back at settlement; the meter is the record of what it took.
An hour of inference a day
| What | Quantity | Rate | Metered |
|---|---|---|---|
| 30 GPU-hours | 108,000 GPU-second | $0.36 per GPU-hour | $10.80 |
| A month, metered | $10.80 | ||
On machines you own, each of these is charged and then credited back at settlement — so the cash figure is zero and the number above is a measurement, not an invoice. It is shown so the scale of the meter is legible.
| Provider | Bills on |
|---|---|
| OpenAI API | input tokens, output tokens, per model |
| Amazon Bedrock | tokens per model, provisioned throughput |
| Together AI | tokens per model, dedicated GPU hours |
| Replicate | GPU seconds, per prediction |
What each provider charges per unit changes often; their own pricing pages are the source.