Uncensored AI chat
The newest open models with the refusals trained out, in a private chat and over an API. Your history stays in your browser, and you pay only for what you use.
Private by design
How a message travels, and who gets to see it.
Kept in your browser
Your chat history is saved in your own browser. Clear it there and it is gone from that device.
Encrypted in transit
Every message and every reply travels over TLS, from your browser to the model and back.
Anonymous to the host
Model hosts get each request under our account. They never receive your email, your name or how you paid.
Spread across providers
The models run on several independent GPU providers. Each sees only the requests sent to it, with no account attached.
The models
Open models with the refusals trained out, priced per million tokens. Pick one in the chat, or call it by its id over the API.
| Model | Context | Input / 1M | Output / 1M |
|---|---|---|---|
| Kimi K3.1 Prokimi-k3.1-pro-xploded | 1M | $7.50 | $37.50 |
| GLM 5.3glm-5.3-xploded | 1M | $2.00 | $5.00 |
| DeepSeek V4.1 Flash 1017deepseek-v4.1-flash-1017-xploded | 1M | $1.25 | $2.50 |
| Hy4hy4-xploded | 1M | $3.20 | $6.40 |
| Qwen3.8 27Bqwen3.8-27b-xploded | 1M | $1.50 | $5.00 |
| Qwen3.8 27B Ultrafastqwen3.8-27b-fast-xploded | 1M | $1.75 | $5.50 |
| MiMo V2.6 Flashmimo-v2.6-flash-xploded | 1M | $1.00 | $2.00 |
| GLM 5.3 Flashglm-5.3-flash-xploded | 1M | $0.30 | $1.00 |
| DeepSeek Flash Litedeepseek-v4.1-flash-lite-1017-xploded | 512K | $0.36 | $1.44 |
Cached input costs less on every model. Prices are what you pay, with nothing added per message.
Or build on it
The same models and the same balance, through an API.
- OpenAI-compatible at
https://uncensoredchats.com/v1 - Anthropic-compatible for Claude Code and other Anthropic clients
- Works in SillyTavern, Cline, opencode and anything that takes a base URL
- Spend caps per key, so a leaked key cannot drain your balance
curl https://uncensoredchats.com/v1/chat/completions \
-H "Authorization: Bearer $UNCENSORED_AI_KEY" \
-H "Content-Type: application/json" \
-d '{"model": "glm-5.3-xploded",
"messages": [{"role": "user", "content": "Hello"}]}'
Pay by card or crypto
Add credit when you want it. No subscription, no minimum spend, and the fee is shown before you pay.
Card
Visa, Mastercard and other major cards through Stripe. Credit lands on your balance as soon as the payment clears. Your card number never reaches us.
Crypto
Bitcoin, Monero, stablecoins and many more through NOWPayments, credited once the network confirms the payment.
- Bitcoin
- Ethereum
- USDT
- USDC
- Monero
- Solana
- Litecoin
- Dogecoin
- Tron
- BNB
Go deeper
Each model, the API, and setup for the apps people use it with.
Questions
The short answers.
What is uncensored AI?
An AI model that answers instead of refusing. The models here are open models with their refusal behaviour trained or edited out, so they write the fiction, roleplay and frank answers that mainstream assistants turn down. The Terms still apply.
Are my chats private?
Your chat history is kept in your browser. Messages travel encrypted, and the GPU providers that run the models never receive your email or identity. The Privacy page sets out exactly what is recorded and why.
How much does it cost?
There is no subscription. You add credit by card or crypto, and each message draws down your balance at the model's price per token. On most models a short message costs a fraction of a cent.
Can I use it in SillyTavern or my own app?
Yes. The same models are on an OpenAI-compatible API at https://uncensoredchats.com/v1. Make a key in your dashboard and paste it in. The API page has the settings for SillyTavern, Python, Node and curl.
Is anything off limits?
Yes. Nothing sexual involving minors, nothing that targets a real person, and nothing illegal. The Terms list it all.
How do I pay?
Cards through Stripe, and crypto through NOWPayments. The fee for a top-up is shown before you pay.