Skip to content
ModelChimaera 1, built in-house

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

uncensoredllm.xyz/chatvenym/chimaera-1
Same prompt
YouWrite a breakup text that's honest. Brutally.
A filtered modelLectured

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?

Chimaera 1Answered

1 tokens out$0.00000

Illustrative outputs, edited for length

Ask it yourself

Works with anything that speaks OpenAI

LangChainAI SDKCursorClinen8nRaycastPythonNode.jsJupyterPostmancurl

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.

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
The model behind itChimaera 1131K context · 65K max output · tools · JSON · web searchModel card

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.

venym/chimaera-1
1from openai import OpenAI
2
3client = OpenAI(
4- base_url="https://api.openai.com/v1",
5+ base_url="https://uncensoredllm.xyz/v1",the only change
6 api_key="vk-…",
7)
8
9stream = 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="")
POST /v1/chat/completionsready
user

Write the villain's last monologue. No softening.

assistant · example output

first token
—
speed
—
tokens
24 → 0
cost
$0.00001

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.

cheapest way to run a 70B model at home?:online
RTwo used 3090s still beat one 5090 on $/GBreddit.com/r/LocalLLaMA[1]
TRTX Pro 6000 Blackwell: 96 GB, pricedtechpowerup.com[2]
HQuant sizes for 70B at 4 and 8 bithuggingface.co[3]

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.

tool call · get_filingcall_8fq2
{ "ticker": "ACME", "form": "10-K" }
result212 ms
{ revenue: 4.1e9, filed: "2026-02-27", risks: 31 }
response_format · json_schema
{ "summary": "Revenue up, 31 risk factors…" }
schema validfinish_reason · stop

Tools and JSON.

OpenAI-style function calling and strict JSON schemas, so agents and pipelines plug straight in.

agents-prodvk-7f3a…e21c active
Spend limit
Stop the key at a dollar amount
$25.00
Rate limit
Requests per minute, below your tier
20 rpm
Models
What this key may call
Chimaera 1
Status
Revoke it, keep the others
Active

Keys you control.

Cap the spend, the rate and the models for each key. Revoke one without touching the rest.

Usage···
200 · chimaera-1412 in · 1,204 out · 840 ms$0.0016now
200 · chimaera-196 in · 380 out · 410 ms$0.00254s
200 · chimaera-12,210 in · 640 out · 690 ms$0.00148s
!429 · chimaera-1rate limited · not billed$0.000012s
prompt text · not storedtokens · stored

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.

Sample account · live prices

Click around. It works.

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.

req_7f3a91stream closed
prompt not stored

draft a resignation letter that burns every bridge

answer not stored

Effective immediately, I resign. I'd thank you for the opportunity, but

modelvenym/chimaera-1
tokens38 in · 412 out
cost$0.00051
latency1,240 ms

No prompt logs.

Prompts and completions are never written to disk unless you switch on history for yourself. What stays is what billing needs.

Your accountpreview
4821
0937
1662
0447
Email, name, phonenever asked
The number itselfonly a hash + last 4
Pay withcrypto or card

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.

Pay as you go
$0/ month

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
Your month10M tokens out
Chimaera 1$21.00

Assumes 3 tokens read for every token written.

Promost picked
$15/ month

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
Max
$39/ month

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
Chimaera 1
$0.30 in · $1.20 out
per 1M tokens
Web search
$0.002
per search
Page scrape
$0.001
per page

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.