Uncensored API. Real tokens. No refusals.
Deep API pricing is straightforward: you pay only for the tokens you actually use, with no hidden fees or monthly subscriptions. Our prepaid model ensures your costs are strictly bounded by the credit you load via crypto.
No subscription. Prepaid credit never expires.
If your bot handles 100,000 requests daily, each with 1,000 input tokens and 500 output tokens, you consume 100M input tokens and 50M output tokens per day. Input cost: 100 * $0.25 = $25. Output cost: 50 * $1.00 = $50. Total daily cost: $75. With prepaid credit, you know exactly how many days your balance will last.
Processing 10,000 stories with 2,000 input tokens and 8,000 output tokens each results in 20M input tokens and 80M output tokens. Input cost: 20 * $0.25 = $5. Output cost: 80 * $1.00 = $80. Total batch cost: $85. Since errors are free, any failed generation doesn't add to this total.
Sending 500 requests with 10,000 input tokens and 1,000 output tokens each uses 5M input tokens and 0.5M output tokens. Input cost: 5 * $0.25 = $1.25. Output cost: 0.5 * $1.00 = $0.50. Total test cost: $1.75. You can budget precisely for high-volume testing without risking subscription overages.
The uncensored deepseek model charges $0.25 per 1 million input tokens and $1.00 per 1 million output tokens. Input tokens cover the prompt you send, including context history. Output tokens cover the model's response. If you do not set a max_tokens limit, the model defaults to 2,048 output tokens per request. This pricing structure is transparent and applies to every API call, ensuring you know exactly what each character of generated text costs.
We accept only cryptocurrency for top-ups: USDT on the TRC20 network or USDC on the Base network. You can load any whole amount between $10 and $500 per transaction. There are no credit cards, no PayPal, and no bank transfers required. This method ensures fast processing and minimal fees. Your prepaid credit is added to your account immediately after confirmation, allowing you to start generating content without delay.
To reward larger loads, we offer bonus credit on top of your deposit. Top-ups of $50 or more receive a +5% bonus. Loads of $100 or more receive a +10% bonus. This bonus credit is applied instantly to your balance. For example, a $100 load gives you $110 in usable credit. This effectively lowers your per-token cost for high-volume users who prefer to preload their accounts.
No tiers: every key gets the full feature set and the same limits.
| Item | Value |
|---|---|
| JSON mode | response_format: {"type": "json_object"} |
| Completion length | prompt + completion fit within 100,000 tokens; max_tokens optional, no separate output cap |
| Function calling | Yes — tools, tool_choice; replies carry tool_calls, also when streaming; send results back as role: tool |
| Max context | 100,000 tokens, input and output combined |
| Parallel requests | up to 8 in parallel per key |
| Requests per minute | 300/min per key |
| Bonus credit | +5% from $50, +10% from $100 |
| Subscription | no monthly fee; paid credit does not expire |
| Token prices | $0.25 per 1M input tokens · $1.00 per 1M output tokens |
| Free trial | $0.50 for 7 days, no card · Trial key: 2 parallel requests, 60 req/min; full limits (8 and 300) after first top-up |
| Payment | USDT (TRC20) or USDC (Base), any whole amount from $10 to $500 |
No. Every new account receives $0.50 in trial credit valid for 7 days. No card is needed to sign up or start your first request. You can pay for future usage entirely with crypto.
No. Once you load credit, it never expires. You can top up whenever you want, and the balance remains available indefinitely until you use it.
Credit is not refunded, but errors are fixed through our Support page. If a double charge occurs or a request fails unexpectedly, contact us to resolve the issue. Failed requests do not consume tokens.
Generating text requires significantly more compute power than reading your prompt. The $1.00 per 1M output tokens reflects the higher GPU cost for producing content. This is standard for most LLM APIs, including uncensored variants.
Create an account, copy the key, change the base URL. That is the whole setup.
