Transparent Token Pricing

Per-Token Rates & Cost Breakdown

Our API charges strictly per token consumed, with input tokens priced at $0.15 per million and output tokens at $0.50 per million. This structure ensures you pay exactly for the computational work performed, without paying for idle time or reserved capacity. The pricing model is straightforward: divide your token count by one million, multiply by the respective rate, and sum the results. Input tokens are cheaper because they represent the data you provide, while output tokens reflect the generative effort of the model. Errors and refusals incur zero cost, meaning you only pay for successful completions. This clarity allows you to predict expenses accurately based on your expected token volume. Unlike providers that bundle features, we isolate the core function: text generation. You can adjust your temperature or top_p settings without affecting the per-token rate. The cost remains linear relative to usage, making scaling predictable. We do not charge for API calls themselves, only for the tokens processed. This approach benefits developers who experience variable traffic patterns, as you avoid paying for unused capacity during low-usage periods.

Crypto Top-up Tiers & Bonuses

We support top-ups exclusively via cryptocurrency, accepting USDT on the TRC20 network and USDC on the Base network. You can deposit any whole amount between $20 and $500 per transaction, providing flexibility for small-scale testing or large-scale deployment. Our bonus structure rewards larger deposits: adding $50 grants a 5% bonus, $100 yields a 10% bonus, and $200 provides a 15% bonus to your credit balance. These bonuses are applied immediately upon confirmation, increasing your usable tokens without additional cost. Your credit never expires, so you can top up whenever convenient. The prepaid nature means your spending is capped by your balance, preventing unexpected bills. This model suits developers who prefer transparent, upfront payment without monthly commitments. You can monitor your remaining credit in real-time through the dashboard. Crypto payments also reduce friction for international developers, avoiding currency conversion fees often associated with traditional payment methods. The system is automated, so your credit is available immediately after network confirmation.

No Subscription, No Hidden Fees

Our service operates on a pure prepaid basis with no monthly subscriptions or recurring charges. You fund your account once and use the credit until it is depleted, at which point you can top up again. There are no hidden fees for API access, rate limits, or feature availability. Errors, refusals, and failed requests do not consume your tokens, ensuring you pay only for valid completions. This model eliminates the risk of paying for unused capacity, which is common with subscription-based APIs. You retain full control over your spending, as credit is tied to your account and does not expire. The absence of subscriptions means you can start small and scale up without contractual obligations. This approach is ideal for developers who want to test the API with a minimal deposit before committing to larger volumes. We do not charge for additional features like streaming or function calling; these are included in the base token rate. The transparency extends to our billing dashboard, where you can view exact token usage and costs. This simplicity reduces administrative overhead, allowing you to focus on integration rather than billing management. Your credit is reserved for your account, ensuring consistency across all your development environments.

Understanding Limits & Errors

Your API key is limited to 300 requests per minute and 8 concurrent requests, with a maximum request body size of 8 MB. These limits ensure stable performance for all users and prevent resource exhaustion. If you exceed the rate limit, you will receive a standard error response, which does not consume your token credit. Errors such as network timeouts or malformed requests also do not deduct from your balance, providing a fair usage model. Each account holds one active key; generating a new key replaces the previous one, so keep your key secure. The prepaid balance remains unaffected by these limits, allowing you to monitor your spend accurately. If you encounter a double charge or billing issue, you can resolve it through our Support page, as credits are non-refundable but adjustable. The 100k context window allows for substantial input, but remember that output tokens are capped at 16,000 per request. Understanding these limits helps you design efficient workflows, such as batching requests or adjusting token limits. Our transparent billing ensures that every token counted is visible in your usage logs, providing complete auditability for your projects.