Pricing you can explain.
See prices before you call, then follow measured usage and the final charge under one Request ID.
01 · Price
Each request retains the price that applied when it ran.
02 · Usage
Cached input, uncached input and output are measured separately.
03 · Review
The final charge is available under the same Request ID.
Current prices
Rates per one million tokens.
| Model | Cached input | Uncached input | Output |
|---|---|---|---|
| Claude Fable 5 claude-fable-5 | $1.00 | $10.00 | $50.00 |
| Claude Fable 5 claude-fable-5-1 | $1.00 | $10.00 | $50.00 |
| Claude Opus 5 claude-opus-5 | $0.50 | $5.00 | $25.00 |
| Claude Opus 5 claude-opus-5-1 | $0.50 | $5.00 | $25.00 |
| Claude Sonnet 4.6 claude-sonnet-4-6 | $0.30 | $3.00 | $15.00 |
| Claude Sonnet 4.6 claude-sonnet-4-6-1 | $0.30 | $3.00 | $15.00 |
| DeepSeek V4 Flash deepseek-v4-flash | $0.0028 | $0.14 | $0.28 |
| DeepSeek V4 Flash deepseek-v4-flash-1 | $0.0028 | $0.14 | $0.28 |
| DeepSeek V4 Flash deepseek-v4-flash-2 | $0.0028 | $0.14 | $0.28 |
| DeepSeek V4 Pro deepseek-v4-pro | $0.003625 | $0.435 | $0.87 |
| DeepSeek V4 Pro deepseek-v4-pro-1 | $0.003625 | $0.435 | $0.87 |
| GLM 5.2 glm-5.2 | $0.26 | $1.40 | $4.40 |
| GLM 5.2 glm-5.2-1 | $0.26 | $1.40 | $4.40 |
| GLM 5.2 glm-5.2-2 | $0.26 | $1.40 | $4.40 |
| GPT-5.6 Luna gpt-5-6-luna | $0.10 | $1.00 | $6.00 |
| GPT-5.6 Luna gpt-5-6-luna-1 | $0.10 | $1.00 | $6.00 |
| GPT-5.6 Sol gpt-5-6-sol | $0.50 | $5.00 | $30.00 |
| GPT-5.6 Sol gpt-5-6-sol-1 | $0.50 | $5.00 | $30.00 |
| GPT-5.6 Terra gpt-5-6-terra | $0.25 | $2.50 | $15.00 |
| GPT-5.6 Terra gpt-5-6-terra-1 | $0.25 | $2.50 | $15.00 |
| Kimi K3 kimi-k3 | $0.30 | $3.00 | $15.00 |
| Kimi K3 kimi-k3-1 | $0.30 | $3.00 | $15.00 |
| MiniMax M3 minimax-m3 | $0.12 | $0.60 | $2.40 |
| MiniMax M3 minimax-m3-1 | $0.12 | $0.60 | $2.40 |
Credits
A positive balance starts a request; actual usage determines the final charge.
Before a call
A positive Team balance is required to start a new request.
At settlement
Requests settle against actual measured usage, including a failed response that consumed tokens. A request may leave the balance below zero; additional calls wait until Credits are added.
One request, one charge
Each Request ID with reported usage produces at most one charge, including when requests settle concurrently.
Pricing FAQ
How is one request charged?
PAI applies the published rates to uncached input, cached input and output tokens separately. The final amount is the exact sum of those three charges.
Can automatic routing change my price?
No. Each model has one published price. Internal route selection and fallback do not change the rate applied to your request.
Can I buy Credits online?
Self-service purchases are not currently available. Your available Credits are shown in Console.