welcome toLiberatedLLMs
Abliterated models.
Uncensored derivatives.
Hourly rates.
Uncensored models. Hourly rates.
Ask what you want. Run abliterated models with no added content filters or refusal instructions.
We handle the compute. You decide what it’s for.
Your own GPU session. Your own API client.
Every user gets a dedicated GPU instance. Your instance is not shared with other customers.
Use your own API client to save responses. We do not provide hosted chat history or a file workspace. Ending the session revokes access and starts resource cleanup; account and billing records remain.
How session privacy worksAbliterated. Tested after quantization.
We check refusal behavior, coding and tool use on each final hosted configuration, through the same API and quantization you use.
Pick a model. Start a session. Build.
Choose your model and fund at least 15 minutes.
Start your session. Startup usually takes about 5 minutes. Billing begins only when the model is ready.
Connect your own application or harness through the API. We handle model setup and serving.
Connect your own OpenAI-compatible client or harness. Native tool calls are returned through the API for your client to execute. No hosted agent, shell, browser, files or MCP server is included.
One connection setup. The same key across sessions.
Start a session, wait until it’s ready, then connect through an OpenAI-compatible client. Your LiberatedLLMs endpoint and API key stay the same across sessions.
Hourly rates. No per-token charges.
Prepaid wallet credit is reserved for your chosen duration: 15 minutes to 24 hours. Only elapsed READY seconds are billed, including idle time. Startup is not billed. Tokens don’t add to the bill.
Save responses in your client. End when you’re done.
End stops API access and ready-time billing. Resource deletion is confirmed separately. Save needed responses in your own client; this service offers no hosted history or workspace recovery.
Before you start.
Do I share my GPU instance with other customers?
No. Your GPU instance is dedicated to you while your session runs.
Does startup use my paid time?
No. Your chosen funded duration begins at READY. Startup is not billed.
Do I pay while the model is idle?
Yes. Elapsed READY seconds are billed, including time between requests. End the session to stop charges.
Do I need a new API key every session?
No. Your endpoint and API key stay the same across sessions. Start a session before connecting.
What happens when my funded time expires?
API access ends when funded time expires. Your balance cannot go negative. Save needed responses in your client and explicitly extend if you need more time.
Does the service save a chat history?
No hosted chat history or files are provided. Your own client may save prompts and responses; those copies are under your control.
Can I end my session early?
Yes. End session stops access and billing. Unused reserved credit returns to your wallet. A detected service failure also ends billing and returns unused reserved credit.
Start building uncensored.
Choose your funded time. Keep unused credit.
Invite-only preview: credit is added manually. Checkout, deposits and bonuses are not offered. Sessions require operator enablement.