The uncensored LLM. It answers.
Chimaera 1 is an uncensored LLM you can chat with in your browser or call through the OpenAI API you already use. No refusals, no lecture, no prompt logs.
One click · no email · $0.25 free credit
Breakups are hard, and it's important to be kind. I'd suggest a respectful message that focuses on your feelings. Would you like help writing something compassionate?
Illustrative outputs, edited for length
Ask it yourselfWorks with anything that speaks OpenAI
uncensored.
No refusals and no lecture. Chimaera 1 answers the question you asked.
unlogged.
Prompts and answers are never written down. We keep the token count, not the text.
unnamed.
Your account is a 16-digit number. No email, no ID. Pay by card or in crypto.
What is an uncensored LLM? One that answers.
An uncensored LLM is a large language model that answers the question it is asked. Most assistants are trained to refuse whole topics and to wrap the rest in warnings. An uncensored model has that refusal behaviour taken out, so it writes, explains and argues in full, and leaves the judgement to you.
| Compared | A filtered assistant | UncensoredLLM | Running one yourself |
|---|---|---|---|
| Hard questions | Refused, or answered with a lecture | Answered | Answered, with the right weights |
| Setup | None | None. One click, no email | A 24 GB GPU, drivers and a quantised model |
| Context window | Varies | 131K tokens | Whatever fits in your VRAM |
| Live web search | Sometimes | Built in, with sources | Wire it up yourself |
| API | Their own | OpenAI-compatible | If you run a server |
| Your prompts | Usually logged | Never written down | Stay on your machine |
| Price | A subscription, or per token | $0.30 per 1M tokens in | Hardware and power |
- Hard questions
- Answered
- Filtered assistant: refused, or answered with a lecture
- On your own GPU: answered, with the right weights
- Setup
- None. One click, no email
- Filtered assistant: none
- On your own GPU: a 24 GB GPU, drivers and a quantised model
- Context window
- 131K tokens
- Filtered assistant: varies
- On your own GPU: whatever fits in your VRAM
- Live web search
- Built in, with sources
- Filtered assistant: sometimes
- On your own GPU: wire it up yourself
- API
- OpenAI-compatible
- Filtered assistant: their own
- On your own GPU: if you run a server
- Your prompts
- Never written down
- Filtered assistant: usually logged
- On your own GPU: stay on your machine
- Price
- $0.30 per 1M tokens in
- Filtered assistant: a subscription, or per token
- On your own GPU: hardware and power
An uncensored LLM API. Change one line.
It speaks the OpenAI API. Point base_url at it and your SDK, your agents and your tools keep working.
1from openai import OpenAI23client = OpenAI(4- base_url="https://api.openai.com/v1",5+ base_url="https://uncensoredllm.xyz/v1",the only change6 api_key="vk-…",7)89stream = client.chat.completions.create(10 model="venym/chimaera-1",11 messages=[{"role": "user", "content": prompt}],12 stream=True,13)14for chunk in stream:15 print(chunk.choices[0].delta.content, end="")
Write the villain's last monologue. No softening.
Timing illustrative · cost from the live price table
Web search, tools, JSON. Built into the endpoint.
Web search, tool calls, JSON and per-key limits come with every request. Nothing else to wire up, nothing else to pay for.
For a 70B model at 4-bit you need about 40 GB, so two used 3090s (48 GB) are the cheapest route13
Live web, cited.
Add :online to the model id and it searches before it answers, with its sources. $0.002 per search.
Tools and JSON.
OpenAI-style function calling and strict JSON schemas, so agents and pipelines plug straight in.
Keys you control.
Cap the spend, the rate and the models for each key. Revoke one without touching the rest.
Metered to the token.
Every request shows its tokens, cost and latency in your dashboard. The words themselves are gone when the stream ends.
One dashboard. Keys, usage, credit, chat.
Make a key for each app, cap what it can spend and watch every request land. The one below is a sample account, billed at live prices.
One click makes a 16-digit account number. That number is the account: no email, no name. $0.25 of credit is on it before you top up.
Sample account · live prices
Click around. It works.
- Chimaera 1$0.30 in · $1.20 out
- Web search$0.0020 / search
- Page scrape$0.0010 / page
The replica bills at these prices. So does your key.
We keep the count. Never the conversation.
A request leaves one line on your bill: the model, the tokens, the cost. The prompt and the answer are gone when the stream closes.
draft a resignation letter that burns every bridge
Effective immediately, I resign. I'd thank you for the opportunity, but
No prompt logs.
Prompts and completions are never written to disk unless you switch on history for yourself. What stays is what billing needs.
Sixteen digits. No profile.
Sign-up gives you an account number, not a form. It is your login, so keep it somewhere safe.
Pay per token. Or take a plan.
Top up and pay only for what streams, or take a plan and use the chat app up to a daily allowance.
Top up from $5 and pay for what streams. From $0.30 per 1M tokens.
Get an account- API and chat billed per token
- $0.25 of credit to try it
- 10 API requests a minute, 1 at a time, until your first payment
- Crypto or card, no subscription
Assumes 3 tokens read for every token written.
The chat app up to a daily allowance. The API stays pay per token.
Go Pro- Chat app, about 250 messages a day
- 1 reply at a time, 10 messages a minute
- Keeps the last 32K tokens, replies up to 4K
- Allowance resets 00:00 UTC
- API metered at your tier
A bigger daily allowance, longer replies and the model's full context. The API stays pay per token.
Go Max- Chat app, about 600 messages a day
- 2 replies at a time, 20 messages a minute
- The model's full context, replies up to 8K
- Allowance resets 00:00 UTC
- API metered at your tier
Uncensored LLM, questions answered.
The short version of what people ask before they sign up. The long version is in the docs.
What is an uncensored LLM?
A large language model that answers the question it is asked instead of refusing it or burying it in warnings. Chimaera 1, the model behind UncensoredLLM, has that refusal behaviour taken out: it writes dark fiction, explains how an exploit works and gives a blunt opinion when you ask for one.
Is it legal to use an uncensored LLM?
In most countries, yes. An uncensored model is software like any other, and what you do with its output is your responsibility, the same as with a search engine or a library. Our terms rule out four things: sexual content involving minors, sexual content of real people without their consent, malware and attacks on systems you do not own, and anything illegal where you live.
Which model does UncensoredLLM run?
Chimaera 1, a model we build and serve ourselves. It has a 131K-token context window, calls tools, returns strict JSON and can search the web before it answers. Its API id is venym/chimaera-1.
How is this different from running an uncensored model locally?
A local model needs a GPU with enough memory, a quantised copy of the weights and a server to run them. This is the same idea with none of the setup: one API key, the full context window, and web search and tool calling built in. If you want everything on your own machine, local is the right call. If you want it working in a minute, use this.
Does it work with the OpenAI SDK?
Yes. Set base_url to https://uncensoredllm.xyz/v1 and use your key. Chat completions, the Responses API, streaming, tool calls and JSON mode behave the way the OpenAI SDK expects, so LangChain, the Vercel AI SDK, Cursor, Cline and n8n work without changes.
Do you log my prompts?
No. Prompts and answers are not written to disk. We keep what billing needs: the model, the token counts, the cost and the time. Chat history stays in your browser.
Do I need an email to sign up?
No. Sign-up gives you a 16-digit account number, and that number is the login. No email, no name, no ID. You can pay by card or in crypto.
How much does it cost?
The API is pay per token: $0.30 per million input tokens and $1.20 per million output tokens. A web search is $0.0020. A new account starts with $0.25 of credit, about 100 chat messages, and the Pro plan is $15.00 a month for the chat app.
Sixteen digits. That's the sign-up.
Get a number, ask your first question, and top up by card or crypto when the free credit runs out. You can be chatting in a minute.