Hermes uncensored uncensored ai API pricing
Hermes uncensored provides a transparent, prepaid billing model for your uncensored AI API needs. You pay only for the tokens you use, with no hidden monthly fees or subscription locks.
- $0.15
- per 1M input tokens
- $0.50
- Output tokens / 1M
- 100,000
- token context
- $0.50
- trial credit
- 300
- requests per minute
No recurring fees. Prepaid balance does not expire.
Usage estimator
Usage examples
- 01
A developer testing a 3,000-token system prompt with a 500-token response.
The cost is calculated as (3,000 / 1,000,000 * 0.15) + (500 / 1,000,000 * 0.50). This equals $0.00045 + $0.00025, totaling $0.0007 per request. Volume scaling makes this negligible for testing loops.
- 02
A content generator running a 50,000-token context window for a 4,000-token output.
Input costs (50,000 / 1,000,000 * 0.15) equal $0.0075. Output costs (4,000 / 1,000,000 * 0.50) equal $0.002. The total request cost is $0.0095. High throughput remains cheap even with large contexts.
- 03
An agent performing 100 tool calls with small 200-token inputs and 100-token outputs each.
Each call costs (200 / 1,000,000 * 0.15) + (100 / 1,000,000 * 0.50) = $0.00003 + $0.00005 = $0.00008. For 100 calls, the total is $0.008. Predictable prepaid credit covers these micro-transactions efficiently.
How prepaid credit works
The uncensored LLM API uses a prepaid balance model. You deposit funds once, and credit deducts based on real token usage. Errors and refusals do not charge you. Your balance never expires, so you can top up when convenient. This approach keeps costs predictable for long-running projects. You avoid the surprise of monthly subscription renewals. The balance tracks input and output tokens separately at their respective rates. You always know exactly what you have left to spend.
Try it with trial credit
New accounts receive $0.50 in trial credit for seven days. This allows you to verify the uncensored AI API key works with your stack. You can test streaming, tools, and JSON mode risk-free. The trial covers enough requests to confirm latency and output quality. If you like the service, you can upgrade to a full prepaid balance. The trial is one per person, ensuring fair access for all developers.
Why prepaid keeps spend predictable
A prepaid balance prevents budget overruns on your uncensored AI models. You set a hard limit by topping up a specific amount. When the balance hits zero, requests stop immediately. This is ideal for VPS deployments or automated agents. Crypto top-ups are instant. You retain full control over your spending without vendor lock-in. Predictability matters when margins are thin.
FAQ
What is the uncensored AI API pricing per token?
The uncensored AI API costs $0.15 per million input tokens and $0.50 per million output tokens. Errors are free. You only pay for tokens actually consumed by the model. This rate applies to all requests regardless of volume. No hidden fees exist for streaming or JSON mode.
How do I top up my credit?
Top up via crypto using USDT (TRC20) or USDC (Base). Minimum and maximum amounts are $20 to $500 per transaction. Bonuses apply automatically: +5% from $50, +10% from $100, and +15% from $200. Payments are processed instantly. You can buy any whole dollar amount within the range.
Does my credit expire?
Prepaid credit never expires. You can keep a balance for years. The $0.50 trial credit expires after seven days. Regular topped-up balance remains active until you use it. This allows you to top up during sales or when funds are available. You retain full ownership of your remaining balance.
Can I get a refund if I top up too much?
Credit is generally non-refundable. If you experience a double charge or error, use the Support page to fix it. The system adjusts your balance accordingly. You do not need to wait for a bank reversal. Support handles corrections directly within your account. This keeps the process fast and transparent.
Get your API key
Sign up, copy the token, and update your endpoint. Done.
