Token pricing
Pay only for the tokens you use with our prepaid balance. We use a crypto-only billing model that keeps your spend predictable and your credit valid indefinitely.
- per 1M input tokens
- $0.25
- Output tokens / 1M
- $1.00
- token context
- 64,000
- trial credit
- $0.50
- requests per minute
- 300
Worked examples
Long-form content generation with heavy reading
A 40,000-token input prompt generates a 2,000-token response. Input cost: 40,000 / 1,000,000 * $0.25 = $0.01. Output cost: 2,000 / 1,000,000 * $1.00 = $0.002. Total: $0.012.
Short, high-frequency chat interactions
A 500-token input prompt generates a 150-token response. Input cost: 500 / 1,000,000 * $0.25 = $0.000125. Output cost: 150 / 1,000,000 * $1.00 = $0.00015. Total: ~$0.000275.
Streaming with error handling
A 10,000-token input generates 4,000 tokens before an API error. Errors are free. Input cost: 10,000 / 1,000,000 * $0.25 = $0.0025. Output cost: 4,000 / 1,000,000 * $1.00 = $0.004. Total: $0.0065.
Token Pricing Rates
We charge precisely for the text tokens processed. There are no hidden fees for streaming, function calling, or JSON mode. The mistral api billing structure is straightforward: input tokens cost $0.25 per million, and output tokens cost $1.00 per million. This pricing applies to every request, regardless of whether the response is successful or ends in an error. If a request fails due to a timeout or network issue, you are not charged for the tokens processed in that request. This ensures you only pay for actual work done by the uncensored llama or other models hosted on our infrastructure. The rate differential reflects the compute intensity of generating text compared to processing it.
Prepaid Credit Model
Our billing operates on a prepaid credit system. You load funds into your account, and tokens are deducted as they are consumed. This model eliminates the risk of surprise invoices or exceeding a monthly budget. Since credit never expires, you can load funds once and use them over months or years. Errors and refusals do not consume your balance. This makes the deepseek api pricing or similar competitor rates easy to compare, as we offer a transparent, usage-based model without subscription commitments. You can track your exact usage in real-time through the dashboard.
Crypto Top-Ups & Bonuses
We accept USDT on the TRC20 network and USDC on the Base network for all top-ups. There are no credit card processing fees or PayPal transaction costs. You can top up any whole amount between $10 and $500. To encourage larger balances, we offer bonus credit: a 5% bonus on top-ups of $50 or more, and a 10% bonus for top-ups of $100 or more. This bonus is added to your prepaid balance immediately. The lack of traditional payment gateways keeps our operational costs low, allowing us to pass those savings to you in the form of bonuses and competitive token rates.
Questions and answers
Do unused credits expire?
No, prepaid credit never expires. You can top up your account once and use the balance over any period of time. This makes our solution ideal for developers who run sporadic inference tasks or want to avoid monthly subscription deadlines.
What happens if a request errors out?
If a request fails due to a server error, timeout, or network issue, you are not charged for the tokens processed in that request. You only pay for tokens that are successfully generated and returned to you, or processed without error.
How do I add funds to my account?
You can top up your balance using USDT (TRC20) or USDC (Base). Navigate to the billing section, select your preferred cryptocurrency, and send the desired amount. Your credit is updated automatically once the transaction is confirmed on the blockchain.
Is there a minimum top-up amount?
Yes, the minimum top-up amount is $10. You can add any whole dollar amount up to $500 per transaction. Larger amounts may be processed individually or split into multiple transactions if needed.
Your key is one form away
Create an account, copy the key, change the base URL. That is the whole setup.
