Prepaid token pricing
Stop letting subscription pauses kill your workflow. We charge strictly by the token, so you pay for what you use and nothing else.
Pricing examples
Long context analysis
You process a 50,000 token document and generate a 2,000 token summary. The cost is (50,000 * $0.15 / 1,000,000) + (2,000 * $0.50 / 1,000,000) = $0.0075 + $0.001 = $0.0085.
High-volume chat inference
A bot handles 1,000 requests with 1,000 input tokens and 500 output tokens each. Total input: 1,000,000 tokens. Total output: 500,000 tokens. Cost: (1,000,000 * $0.15 / 1,000,000) + (500,000 * $0.50 / 1,000,000) = $0.15 + $0.25 = $0.40.
Code generation burst
You send 4,000 tokens of code context and receive 8,000 tokens of generated logic. Input cost: (4,000 * $0.15 / 1,000,000) = $0.0006. Output cost: (8,000 * $0.50 / 1,000,000) = $0.0040. Total: $0.0046.
Per-Token Rates
Input tokens cost $0.15 per million and output tokens cost $0.50 per million. You only pay for actual usage, not idle time. This transparent structure eliminates the risk of hidden fees that plague other platforms. Your API key tracks every byte sent and received. We charge strictly by the token, ensuring you pay for what you use and nothing else. Errors do not consume credit. Refusals do not charge you. The math is simple and predictable. You can budget your infrastructure costs with precision. This model scales with your volume. Small projects pay cents. Large deployments pay dollars. There are no monthly minimums. There are no surprise charges. You control your spending exactly.
Prepaid Credit & Expiry
Prepaid credit never expires. Your balance remains available indefinitely until you use it. This feature supports long-term projects that may have variable usage patterns. You top up once and use the funds over time. The credit sits in your account securely. You can add funds at any time. The system deducts tokens from your balance in real time. When the balance hits zero, your requests stop. This prevents runaway costs. You decide when to add more. There is no subscription clock ticking down. Your credit is yours. It waits for you whenever you are ready to work.
Crypto Top-Up Bonuses
We accept USDT on TRC20 or USDC on Base for all top-ups. You can deposit any whole amount between $20 and $500. Larger deposits earn bonus credit automatically. A $50 top-up gives you +5% bonus. A $100 deposit gives you +10% bonus. A $200 deposit gives you +15% bonus. These bonuses increase your effective credit instantly. You get more tokens for your crypto. The process is fast. You sign in with Google or email and password. Your key appears immediately. The bonus applies to your total balance. This makes high-volume usage even cheaper. You maximize every dollar you spend.
Common questions
What is the claude NSFW policy for this API?
We do not use Claude. We use an uncensored model that does not refuse lawful adult content. You can generate NSFW text without policy violations. Our model is tuned for unrestricted output. It works like a proxy but runs its own weights. You get consistent results without the risk of your account being paused for content. This is an uncensored LLM designed for developers who need reliability.
Which payment methods do you accept?
Crypto only: USDT on TRC20 or USDC on Base, any whole amount from $20 to $500. Bonus credit applies from $50.
What happens if the API returns an error?
Errors are free. If the server returns a 5xx error, you are not charged for that request. Refusals due to content policies are also free. You only pay for successful completions. This ensures your billing matches your actual utility. The API returns token usage in the last chunk of the stream.
Can I use this for an NSFW chatbot?
Yes. The model is uncensored and handles adult content well. It supports streaming via SSE for real-time responses. You can use standard OpenAI SDKs. The API key works with any compatible client. You get an unrestricted AI experience. It is a cheap LLM API for chatbots that need high volume.
Get your key now
Sign up, copy the key, update the base URL. Setup takes seconds.
