Uncensored LLM API pricing. Up to 30% below list.

Uncensored LLMs are language models tuned or edited to answer instead of refusing or moralizing. Venice Uncensored is fine-tuned for it, while Abliterated Large V2 and GLM-4.7 Flash Heretic have the refusal behavior edited out of their weights. Zurelay serves every one through one OpenAI-compatible endpoint, on prepaid credit with no subscription. Fewer refusals don't change the rules: Zurelay's Terms of Service apply to everything you send.

Uncensored LLM models
5
Below list price
2–30%
From, per 1M input
$0.061
Get your API keySee every price
PlaygroundDemo

Chat with Venice Uncensored 1.2

Pick a prompt to see a reply stream in, priced at Venice Uncensored 1.2’s rate.

Sample replies, priced live. Nothing is sent.

Pricing

Every uncensored LLM model. Live prices, one key.

Uncensored LLM API pricing, in short: List prices for uncensored LLM models are $0.07 to $3.00 per 1M input tokens and $0.40 to $5.00 per 1M output tokens. On Zurelay the same models cost $0.061 to $2.93 input and $0.35 to $4.88 output, 2–30% less, with no subscription.

Uncensored LLM API pricing: list price and Zurelay's, per 1M tokens
List priceZurelay
ModelInput / 1MOutput / 1MInput / 1MOutput / 1MYou save
Abliterated Large V2abliterated-large-v2$3.00$5.00$2.93$4.882%off
Venice Uncensored 1.2venice-uncensored-1.2$0.20$0.90$0.14$0.6330%off
Venice Uncensored Role-Playvenice-uncensored-role-play$0.50$2.00$0.35$1.4030%off
GLM-4.7 Flash Hereticglm-4.7-flash-heretic$0.07$0.40$0.061$0.3513%off
Gemma 4 Uncensoredgemma-4-uncensored$0.1625$0.50$0.155$0.484%off

Prices in US dollars per 1M tokens, live from the catalog. List prices are each model's own published API price. You save is on output tokens; each model's page has its cached-input price.

Which model

Which model to use. Same key for all of them.

30%off

Creative writing and frank chat

Venice Uncensored 1.2

$0.14 / $0.63 per 1M tokens

Venice's uncensored chat model, a fine-tune of Mistral's 24B, for fiction, brainstorming and direct answers on sensitive topics, with image input and a 128K-token context.

Venice Uncensored 1.2 API
30%off

Roleplay and characters

Venice Uncensored Role-Play

$0.35 / $1.40 per 1M tokens

Venice's uncensored model tuned to follow a persona from your system prompt and stay in character, for companions and interactive fiction.

Venice Uncensored Role-Play API
2%off

Security research and red-teaming

Abliterated Large V2

$2.93 / $4.88 per 1M tokens

Abliteration AI's large model with its refusals removed, which its docs link to Z.ai's GLM-5.3, with a 1M-token context.

Abliterated Large V2 API
13%off

Low-cost reasoning and tools

GLM-4.7 Flash Heretic

$0.061 / $0.35 per 1M tokens

A community build of Z.ai's GLM-4.7 Flash with its refusals removed, with reasoning you can switch off, tool calling and a 200K-token context.

GLM-4.7 Flash Heretic API
4%off

Images and tools

Gemma 4 Uncensored

$0.155 / $0.48 per 1M tokens

A community uncensored fine-tune of Google's Gemma 4 26B, with image input, tool calling and a 256K-token context.

Gemma 4 Uncensored API

What it means

Uncensored, explained. Two ways to get there.

An uncensored model answers instead of refusing or lecturing. There are two common ways to make one. Fine-tuning trains a model further on examples where it answers, which is how Venice builds Venice Uncensored and Venice Uncensored Role-Play. Abliteration edits the weights instead: it compares the model's internal activations on prompts it refuses and prompts it answers, finds the direction linked to refusing and suppresses it, without retraining. Abliterated Large V2 and GLM-4.7 Flash Heretic are made this way.

Removing refusals can shift other abilities too, such as instruction following, tool use or code quality, so test a model on your own tasks before you rely on it. Community builds such as GLM-4.7 Flash Heretic and Gemma 4 Uncensored aren't made or endorsed by the labs whose open weights they start from.

The rules

Fewer refusals. Same rules.

An uncensored model drops the refusals, not the rules. Zurelay's Terms of Service apply to everything sent to these models, as to every other model, and you're responsible for how you use what they write. Section 7 of the Terms still rules out, among other things:

  • Child sexual abuse material or anything that sexualizes minors, which we report to the National Center for Missing & Exploited Children and to law enforcement
  • Help with weapons capable of mass casualties, terrorism or violent extremism
  • Malware, phishing and attacks on systems you aren't authorized to test
  • Fraud, scams, impersonation and spam
  • Harassment, threats and defamation

Read the Terms of Service

Quickstart

Change two lines. Keep your code.

Point the OpenAI SDK, or any OpenAI-compatible tool, at https://api.zurelay.com/v1 and use your Zurelay key. Streaming, tool calling and structured output work as documented.

Add credit from $10, with no subscription. The quickstart has the rest.

chat.py
from openai import OpenAI
client = OpenAI(
    base_url="https://api.zurelay.com/v1",
    api_key="YOUR_ZURELAY_KEY",
)
stream = client.chat.completions.create(
    model="venice-uncensored-1.2",
    messages=[
        {"role": "user", "content": "Explain quicksort briefly."},
    ],
    stream=True,
)
for chunk in stream:
    if chunk.choices:
        print(chunk.choices[0].delta.content or "", end="")

Verified

The model you asked for. Tested before they serve.

Before a route serves requests, it's tested: the model has to say what it is, answer correctly, stream, and handle the features its page lists, such as images, tools or reasoning levels. Failed attempts aren't billed.

Most of these models have only one or two routes, fewer than the big labs' models, so one can be unavailable for a while. If an answer doesn't look right, send its x-request-id to support and we check the route that served it.

Live health for every model is on the status page.

FAQ

Uncensored LLM API questions.

Anything else? Ask our team and we’ll answer by email.

How much does the uncensored LLM API cost?

List prices for uncensored LLM models are $0.07 to $3.00 per 1M input tokens and $0.40 to $5.00 per 1M output tokens. On Zurelay the same models cost $0.061 to $2.93 input and $0.35 to $4.88 output, 2–30% less, with no subscription. The lowest-priced is GLM-4.7 Flash Heretic at $0.061 input and $0.35 output per 1M tokens on Zurelay; the highest is Abliterated Large V2 at $2.93 and $4.88, 2–30% below list across the family. You add prepaid credit from $10 and pay for the tokens you use.

Are these the real models?

Yes. Each request runs on the model you name, never a smaller or substitute model, and each route is tested before it serves requests. Most of these models have only one or two routes, so one can be unavailable for a while; the status page shows live health, and a failed attempt isn't billed.

Why is it cheaper than list price?

Sellers offer model capacity they bought ahead, through prepaid credits and volume discounts, below list price. Zurelay tests it and sends each request to a route that is available and verified. You pay per token from prepaid credit, with no subscription.

How do I start calling these models through Zurelay?

Create a key, then point the OpenAI SDK, or any OpenAI-compatible tool, at https://api.zurelay.com/v1 with a model ID such as venice-uncensored-1.2. Code that already uses the OpenAI SDK only needs the base URL, the key and the model name changed.

What is an uncensored LLM used for?

Fiction and roleplay with dark or mature themes, brainstorming without guardrails, frank answers on sensitive topics, authorized security research and red-teaming your own AI systems and safety filters. The Terms of Service still apply: no content that sexualizes minors, no help with weapons of mass harm, no malware or attacks on systems you aren't authorized to test, and no fraud.

Are there uncensored image or video models?

Not on Zurelay today. Every uncensored model here is a chat model, called through POST /v1/chat/completions.

More pricing

Stop paying list price.
Start saving today.

Create a key in seconds, point the SDK you already use at Zurelay, and every request costs up to 90% less from the first token.

Get your API keyTalk to an engineer