Pricing
Prepaid credit. Pay only for tokens. No subscriptions.
| Model | Tier | Input / 1M | Cached input / 1M | Output / 1M |
|---|---|---|---|---|
qwen3.8-27b-uncensored | standard | $0.6 | $0.1 | $2.25 |
qwen3.8-27b-uncensored | flex | $0.3 | $0.05 | $1.125 |
How you are billed
- Prompt tokens served from the prefix cache are billed at the cached-input rate. All other prompt tokens are billed at the input rate.
- Output tokens include reasoning tokens.
- Before a request runs, its worst-case cost is held from your balance. The unused part is released when it finishes.
- If you disconnect or cancel, you pay for the tokens generated until the stream stopped.
- Requests that fail because of an engine fault or an invalid model output are not billed.
- The
flextier costs half and runs only on idle capacity. It can be rejected with HTTP 429 when the service is busy.