Transparent token pricing, no hidden fees

Token costs: input vs output

The uncensored model charges $0.15 per million input tokens and $0.50 per million output tokens. This ratio reflects the computational difference between processing your prompt and generating the response. You pay exactly what you consume.

Errors do not cost you. If the API returns an error or a refusal, you are not charged for those tokens. This protects your budget when testing edge cases or running noisy data pipelines.

Every token counts toward your total usage. The 100,000 token context window includes both the input and the output. Plan your requests accordingly to maximize your prepaid balance.

Prepaid credit and crypto payments

We accept crypto only: USDT on the TRC20 network or USDC on Base. Top-ups range from $20 to $500 in whole amounts. Your credit never expires, so you can top up when it suits your workflow.

You get a $0.50 trial credit for new accounts. This lets you verify the API works with your existing SDK before committing funds. The trial is valid for seven days.

Support handles refunds for errors like double charges. Your main balance remains intact. This keeps our overhead low and your costs predictable.

Bonus credit tiers for larger top-ups

Larger top-ups earn you bonus credit. Add $50 and receive a 5% bonus. Add $100 for a 10% bonus. Add $200 and get a 15% bonus. These bonuses are applied immediately to your account.

This structure rewards heavy users without locking them into a subscription. You still pay the same token rates. The bonus simply increases your purchasing power for inference.

Use the bonus credit for long-running agent tasks or batch processing. The credit behaves exactly like your base balance. It expires only if you manually request a refund, which is rare.

Why prepaid is better than subscriptions

Subscriptions lock you into monthly fees even if you stop using the API. Prepaid credit gives you full control. You pay for what you use. If your project pauses, your balance sits there waiting.

No hidden fees. No annual commitments. No surprise overages. You know exactly how many requests your budget covers. This transparency is crucial for developers running agents on constrained hardware.

Competitors bundle features you might not need. We offer a single, high-capacity uncensored model. You get pure token economics. Every cent buys inference. No feature tax. No tier upgrades.

Rate limits and error handling costs

You get 300 requests per minute per key. Eight concurrent requests are allowed. The request body is limited to 8 MB. These limits ensure stable performance for all users.

Errors are free. If a request fails due to rate limits or server issues, you are not charged. This protects your prepaid balance from unexpected costs during high-traffic periods.

Use a single API key per account. If you need a new key, it replaces the old one. This simplifies your integration. You do not need to manage multiple keys for different environments.