Pricing
Pay for the tokens you use.
No subscription and no minimum. Each model has an input price and an output price, and you set the most your workspace can spend in a month.
Per-token prices
Price per one million tokens
In US dollars, excluding applicable taxes.
| Model | Model ID | Input / 1M | Output / 1M | Status |
|---|---|---|---|---|
| Qwen-SEA-LION-v4.5-27B-IT | qwen-sea-lion-v4.5-27b |
$0.60 | $3.60 | Live |
| GPT-6 Astra | gpt-6-astra |
$10.00 | $50.00 | Live |
| Kimi K3 | k3 |
Announced at launch | Coming soon | |
| GLM-5.3 | glm-5.3 |
Announced at launch | Coming soon | |
How billing works
Metered per request, invoiced per month.
-
Every request is metered
The model reports the input and output tokens for each request. Both are recorded with their cost and shown in your usage.
-
Your limit is checked first
Before a request is sent, its largest possible cost is reserved against your monthly limit. A request that would pass the limit is refused.
-
One invoice per month
Usage for the calendar month is invoiced after the month closes and charged to the payment method on your workspace.
Questions
Pricing and billing, in detail
Is there a subscription or a minimum spend?
No. You pay for the tokens your requests use, at the price of the model you call.
What counts as a token?
Input tokens are everything you send in a request, including system and earlier messages. Output tokens are what the model generates. The counts come from the model for each request and appear in your usage.
Am I charged for a request that fails?
A request that fails before the model returns a response is recorded as non-billable.
How does the spending limit work?
You set a monthly limit for your workspace in the console. The limit is checked before each request is sent, so it stops spend at the limit rather than reporting it afterwards. You can change the limit at any time.
How do I pay?
By card. Card details are entered in a form hosted by Stripe and never reach MarsCompute. Tax is calculated from your billing address.
Can I see what each request cost?
Yes. Usage lists requests with their token counts and cost, can be filtered by member, key and model, and can be exported as CSV.
Set your limit, then send a request.
Request access and we will set up your workspace.