Summary
- The model does not refuse lawful adult or controversial topics, but it strictly blocks sexual content involving minors.
- You pay only for the tokens you use, with no monthly subscription or hidden fees attached to your account.
- Crypto top-ups provide bonus credit, and the prepaid balance never expires, making costs predictable.
- The API is OpenAI-compatible, meaning you can switch your client code with just a base URL change.
Myth: It is just a filtered ChatGPT wrapper
It is a distinct model, not a wrapper around ChatGPT. The service runs an open-weight model on its own servers, tuned to answer without content refusals. It is not GPT, Claude, or any other vendor's model. When you send a request, you are talking directly to this uncensored LLM. It does not proxy another company's API. This means you get predictable costs and direct model access. You do not need to worry about another provider changing their filters or prices. Our base URL is HTTPS://api.unfilteredgptchat.com/v1. You can use the official OpenAI SDKs with this endpoint. The model id to send is "uncensored". It handles text in and text out. There are no embeddings, images, or audio. It is a standalone uncensored model. This ensures you are not paying for features you do not use.
Myth: Prompts are used to train the model
Your prompts are not used for training. Privacy is a key feature of this service. When you sign up, you only need an email address. There is no phone number required. The prompts you send are processed to generate a response. They are not stored to improve the model's weights. This means your data remains yours. You can use it for sensitive creative writing or security research. The model does not remember your previous conversations. Each request is processed independently. This is different from some free chat interfaces that use your data for improvement. Our model is tuned for adult use. It is not trained on your specific inputs. This keeps your content private and your experiments repeatable.
Myth: You need a monthly subscription
You do not need a monthly subscription. The pricing is transparent and prepaid. You pay for what you use. The cost is $0.15 per 1M input tokens. Output tokens cost $0.50 per 1M. Errors and refusals are free. You never pay for unused capacity. There are no hidden fees. Your credit never expires. You can top up whenever you want. This is ideal for developers who need a drop-in API. You do not lose your balance if you take a break. The prepaid model allows you to control your spend exactly. You can stop using the API at any time. There is no recurring charge on your bank account. This makes it a cheap LLM API for variable workloads.
Myth: All NSFW content is allowed without limit
There is one hard content limit. Sexual content involving minors is always refused. This is a strict rule that applies to all requests. Other adult content is allowed. The model does not refuse lawful adult topics. It also handles controversial or security-research prompts. Fictional content is fine. The model is tuned for these uses. You can generate text without the usual polite refusals. However, the minor content filter is non-negotiable. This is the only limit that always applies. Everything else is open. This makes it an uncensored model for most use cases. You do not need to guess what is allowed. The rules are clear and consistent.
Myth: It works exactly like Grok or Claude
It is not Grok or Claude. The model is open-weight and runs on our servers. It is not a proxy for those services. When you use this API, you are using our specific uncensored LLM. It is not the same model as Grok. Grok has its own limits and features. You should check Grok's documentation for their current prices. This API is an OpenAI-compatible alternative. It supports function calling and JSON mode. It streams via SSE. The context window is 100,000 tokens. Max output is 16,000 tokens. This is different from other vendors. You get direct access to this model. You do not get routing between multiple models. It is a single, focused service.
Myth: Crypto payments mean slow refunds
Crypto payments are fast and transparent. You can top up with USDT on TRC20 or USDC on Base. The amount must be a whole number from $20 to $500. You get bonus credit with each top-up. A $50 top-up gives +5% bonus credit. A $100 top-up gives +10% bonus. A $200 top-up gives +15% bonus. Your credit never expires. If you make a mistake, like a double charge, you can fix it through the Support page. The credit is not refunded in the traditional sense. It stays in your account. You do not need to wait for clearing. The bonus credit is added immediately. This makes the prepaid model efficient.
Fact: Prepaid tokens never expire
Your credit is yours forever. There is no expiration date. You can top up today and use it next year. This is a key benefit of the prepaid model. You do not lose your balance if you stop using the API. The trial credit is valid for 7 days. Every new account gets $0.50 of trial credit. This lets you test the API before you commit. The prepaid credit is charged by real token usage. You pay only for what you consume. This is the most transparent pricing model available. You can predict your costs exactly. There are no surprises at the end of the month.
Fact: One key replaces the old one
You get one active key per account. If you generate a new key, the old one stops working. This is a security feature. You can manage your keys easily. The key is shown immediately when you sign up. You can sign up with Google or email. There is no phone number required. The key is used for all requests. You do not need multiple keys for different projects. One key is enough. You can regenerate it if you suspect it is compromised. The old key is invalid immediately. This keeps your account secure. The rate limits apply per key. You get 300 requests per minute. This is a strict limit to prevent abuse.
Fact: Strict rate limits prevent abuse
The API has strict rate limits. You get 300 requests per minute per key. You can have 8 requests at the same time. The request body limit is 8 MB. These limits ensure fair use. They prevent any single user from overwhelming the servers. This keeps the service fast for everyone. The limits are per key. If you need more, you can generate a new key. But only one is active at a time. The 8 MB limit is enough for most prompts. The 100,000 token context window fits within this limit. The rate limits are predictable. You can plan your usage around them. This ensures stable performance for your applications.
