Uncensored AI, ready for your agents.
- Modelglm-5.3-flash-uncensored, qwen3.8-flash-next-uncensored
- Context128K
- Tool callingsupported
- Plansfrom Rp 7.000
- Tokensno expiry
1 million tokens · Rp 7.000
What you get with Uncensored from Server Host.
No moral filtering
The model answers without added moral refusals, suited to creative writing, security research, and data analysis.
Native tool calling
OpenAI-format function calling for agents that run terminal commands, read files, and debug code.
128K context
Fit long documents, chat histories, or large parts of a codebase into one request.
Thinking, streaming, structured output
Thinking mode for hard tasks, token-by-token streaming, and structured output for automation.
Works in the tools you use
Cursor, Cline, Roo Code, SillyTavern, and the OpenAI SDKs for Python or JavaScript; just fill in the base URL, API key, and model name.
Tokens never expire
Pay once from a QRIS balance. Every new pack adds to the same API key.
Buy tokens once, use them with no expiry.
The glm-5.3-flash-uncensored, qwen3.8-flash-next-uncensored model without moral filtering, with tool calling for coding agents. Tokens never expire, and every further pack adds to the same key.
- Model glm-5.3-flash-uncensored, qwen3.8-flash-next-uncensored, uncensored
- Tool calling, thinking mode, and streaming
- 128K token context
- Tokens stay active forever, no subscription
- Works in Cursor, Cline, Roo Code, SillyTavern, and the OpenAI SDKs
Paid from your balance. Every order states its intended use and is activated by our team after review.
Frequently asked questions.
What is this model good for?
Autonomous coding agents, multi-agent setups, refactoring and debugging large projects, scraping and data processing, roleplay, and creative writing.
What is still not allowed?
Child sexual exploitation content, cyberattacks on critical infrastructure, chemical, biological, radiological, or nuclear weapons guidance, mass phishing and fraud, and doxxing. Violations end the key for good without a refund.
How are tokens counted?
Input and output tokens both count toward the quota. When it runs out, requests are refused until you buy another pack.
Can tokens expire?
No. Tokens stay active forever, and new packs are added to the remaining quota of the same key.
Why do I have to state my intended use?
This model is uncensored, so our team reviews every order before activation. Orders with an unacceptable use are refunded to your balance.
How do I use it in Cursor, Cline, or Roo Code?
Choose the OpenAI Compatible provider, fill in the base URL and API key from the panel, then enter glm-5.3-flash-uncensored as the model name.
Who is responsible for prompts and answers?
You are, fully. We only record token counts to track the quota; prompts and answers are not stored. Avoid sending passwords or personal data.
Everything an online project needs, in one balance.
Set up your first server today.
Sign up for free. You only pay for the services you order.