---
title: "Uncensy alternative: uncensored LLM API, per token"
canonical: https://unrestrictable.ai/alternatives/uncensy
generated: 2026-11-16T02:57:39.583Z
source: https://unrestrictable.ai/alternatives/uncensy
base_url: https://api.unrestrictable.ai/v1
---

# Uncensy alternative: uncensored model API for developers

Uncensy is a consumer chat app sold by subscription. This is the developer route: open-weight uncensored models by API, one prepaid key, priced per token, in any OpenAI-compatible client.

Published October 27, 2026 · prices as of November 16, 2026

Uncensy and Unrestrictable sell different things. Uncensy is a consumer chat app with a character builder, images, voice and a subscription. Unrestrictable is the model layer under an app like that: open-weight uncensored models sold per token on one prepaid key, for developers who call the API from code or from a client such as SillyTavern or Janitor AI. Uncensy's facts were read on 27 October 2026; every price here is read from the live catalog when the page renders.

## Who should pick which

- Pick Uncensy: for a finished consumer chat app: a character builder, chat with memory, voice calls, generated images and video, all under one plan with a free tier.
- Pick Unrestrictable: for the model without the app: 5 open-weight uncensored builds in the table below, billed per token from a prepaid balance that starts at $10, in any code or client that takes an OpenAI-compatible key.
- Both: serve models without the stock refusal layer. The difference is who owns the prompts, the history and the bill.

## Uncensy in one paragraph

[Uncensy](https://uncensy.com/) is a consumer chat app built around custom AI characters. Its home page, read on 27 October 2026, lists a character builder for appearance, voice and personality, a public character browser, streaming chat with memory, voice messages and live calls, generated images and video, and a marketplace where characters change hands for an in-app currency called diamonds. It names three plans: Free, with a daily message allowance and one character slot; Spark, from $7.99 a month with unlimited chat and five slots; and Blaze, with ten slots and more diamonds. The current figures are in the [pricing section of its home page](https://uncensy.com/#pricing); they change, so this page does not repeat more than that. The page mentions no API.

Uncensy suits a reader who wants the whole product in one tab and a flat monthly charge. The subscription buys the app as much as the model.

## Unrestrictable: open-weight models, priced per token

Unrestrictable sells the other half. There is no app to subscribe to. There is a key, a base URL and a list of open-weight builds without the stock refusal layer: abliterated general models, uncensored fine-tunes and fiction tunes for long-form writing. You bring the client: your own code through the OpenAI SDK, [SillyTavern](https://unrestrictable.ai/docs#sillytavern), [Janitor AI](https://unrestrictable.ai/docs#janitor-ai), or the [interactive fiction chat](https://unrestrictable.ai/roleplay) on this site, which comes with 5 character cards and a few free turns a day before it asks for a key.

Prices are read from the live catalog when the page renders. The builds below cover general chat, coding, research and fiction.

| Model | Build | Input $/M | Output $/M | Context | Wire formats |
| --- | --- | --- | --- | --- | --- |
| [GLM-5.3 Uncensored](https://unrestrictable.ai/models/glm-5.3-uncensored) glm-5.3-uncensored | Uncensored | $0.282 | $0.846 | — | openai |
| [DeepSeek V4.1 Flash Abliterated](https://unrestrictable.ai/models/deepseek-v4.1-flash-abliterated) deepseek-v4.1-flash-abliterated | Abliterated | $0.8488 | $1.6975 | — | openai |
| [Qwen3.8 27B Uncensored](https://unrestrictable.ai/models/qwen3.8-27b-uncensored) qwen3.8-27b-uncensored | Uncensored | $1.2125 | $3.6375 | 33K | openai |
| [Llama 3.3 70B Abliterated](https://unrestrictable.ai/models/llama-3.3-70b-abliterated) llama-3.3-70b-abliterated | Abliterated | $2.80 | $2.80 | 33K | openai |
| [Cydonia 24B v4.3](https://unrestrictable.ai/models/cydonia-24b-v4.3) cydonia-24b-v4.3 | Roleplay | $0.484 | $0.60 | 131K | openai |

Prices per million tokens, read from the live catalog on 2026-11-16.

The whole catalog, abliterated, uncensored, fiction and security builds, is on the [models page](https://unrestrictable.ai/models). Top-ups start at $10, credit lasts 12 months, and a month with no messages costs nothing.

Every model gets the same fixed prompt once a day and the answer is compared with the day before. On 2026-11-16, 321 of 347 checked models were stable. The verdict per model is on the [daily check page](https://unrestrictable.ai/proof).

## One working session, costed

Take one working session: 150 requests, each sending about 3,000 tokens of system prompt and history, each answered in about 250 tokens. A writer drafting chapters with a co-writer prompt sends about that much. So does a test harness probing refusal behaviour. On a subscription that session is part of the monthly charge. Here it is arithmetic, done from the catalog prices on 2026-11-16.

| Model | 150 requests | Per request |
| --- | --- | --- |
| [GLM-5.3 Uncensored](https://unrestrictable.ai/models/glm-5.3-uncensored) | $0.16 | $0.0011 |
| [DeepSeek V4.1 Flash Abliterated](https://unrestrictable.ai/models/deepseek-v4.1-flash-abliterated) | $0.45 | $0.003 |
| [Qwen3.8 27B Uncensored](https://unrestrictable.ai/models/qwen3.8-27b-uncensored) | $0.68 | $0.0045 |
| [Llama 3.3 70B Abliterated](https://unrestrictable.ai/models/llama-3.3-70b-abliterated) | $1.37 | $0.0091 |
| [Cydonia 24B v4.3](https://unrestrictable.ai/models/cydonia-24b-v4.3) | $0.24 | $0.0016 |

Computed at render: requests × 3,000 × input price, plus requests × 250 × output price.

The session runs from $0.16 to $1.37 across these builds. Input is the larger share, because every request resends the system prompt and the history. A shorter prompt moves the bill more than the choice of model. A $10 top-up covers 63 such sessions on the lowest row.

## Feature by feature

- Billing: Uncensy: Free, Spark and Blaze plans (read 27 October 2026). Unrestrictable: per token from a prepaid balance, no plan.
- Prompts and characters: Uncensy: built and kept in its app. Unrestrictable: a system prompt you own, or a character card as SillyTavern JSON, usable in any client.
- Chat history: Uncensy: stored by the app, with memory as a feature. Unrestrictable: kept by your client; the gateway stores no request bodies unless you turn replay on for a key. A chat that outgrows the model's window can drop its middle turns instead of failing (middle-out), and hosted agents keep memory per session.
- Images and voice: Uncensy: photos, video and voice calls in the app. Unrestrictable: text models on this page; uncensored image and video models are a separate list on the media page, billed per image or per second.
- API: Uncensy: none mentioned on its site. Unrestrictable: OpenAI-compatible at https://api.unrestrictable.ai/v1.
- Receipts: Unrestrictable: one usage record per request, a signed receipt and a signed monthly statement.

## How to switch

Create a key in the [console](https://app.unrestrictable.ai/console), top up from $10, and send the first request. The system message sets the task. The rest is a normal chat completion.

```bash
curl https://api.unrestrictable.ai/v1/chat/completions \
  -H "Authorization: Bearer sk-unr-live-…" \
  -H "Content-Type: application/json" \
  -d '{"model": "glm-5.3-uncensored", "messages": [{"role": "system", "content": "You are the co-writer on a crime novel. Write in close third person."}, {"role": "user", "content": "Draft the interrogation scene in chapter 12."}]}'
```

In SillyTavern: API Chat Completion, source Custom (OpenAI-compatible), endpoint `https://api.unrestrictable.ai/v1`, your key, then pick the model. Janitor AI takes `https://api.unrestrictable.ai/v1/chat/completions` as the proxy URL. The [docs page](https://unrestrictable.ai/docs) has the fields for every client, and the [playground](https://unrestrictable.ai/playground) needs no setup at all.

## What each side stores

Unrestrictable keeps, per the [privacy page](https://unrestrictable.ai/privacy): the checkout email and amount as the payment processor reports them, keys as salted hashes, one usage record per request with model, token counts and charge, and SHA-256 fingerprints of each request and response. It does not keep the text of a chat unless you switch replay on for a key, which holds that key's request bodies for 7 days.

Uncensy describes its own storage on its site; it is not summarized here. A chat app has to store the character and the history to offer memory across sessions. On Uncensy the app remembers. Here your client remembers and the gateway does not.

## Where Uncensy is the better pick

- You want images and voice in the same app as the chat.
- You would rather pay one monthly number than watch a meter.
- You do not want to set up a client or write a system prompt.

## Questions

### What is Uncensy?

A consumer AI chat app at uncensy.com: a builder for custom characters, chat with memory and voice, generated images and video, and a character marketplace, sold through Free, Spark and Blaze plans (read 27 October 2026).

### Is there an Uncensy alternative without a subscription?

Unrestrictable: open-weight uncensored models priced per token on a prepaid balance from $10. The session above costs cents, and a quiet month costs nothing. See per-token prices.

### Does Uncensy have an API?

Its home page mentions none as of 27 October 2026. Unrestrictable is the API: OpenAI-compatible, one key, every client that takes a custom base URL. See client setup.

### Can I run uncensored models in my own app?

Yes. Point the OpenAI SDK, SillyTavern or Janitor AI at the base URL, send your key, and pick a model id. The SillyTavern guide covers the endpoint, the context size and the sampler settings. See the SillyTavern guide.

### Does the API filter prompts?

Chat requests are not filtered for topic. The acceptable use policy lists what is never allowed, and image and video prompts involving minors are refused. See the acceptable use policy.

Every model above is on the [uncensored model list](https://unrestrictable.ai/models) with its live price per million tokens. [Per-token pricing](https://unrestrictable.ai/pricing) sets the whole catalog side by side, and the [family directory](https://unrestrictable.ai/uncensored) groups the builds by base model.

[Get a key](https://app.unrestrictable.ai/console) Prepaid from $10, no subscription.

## New models and price changes, once a month.

New uncensored builds with their prices, price moves, and what the daily model check found.
