Private AI
Ask anything. We keep nothing.
A fast chat on open models. Your prompts and answers are never stored and never used for training.
Free to try: 50,000 tokens a day after you sign in.
The first request after a quiet period can take a couple of minutes while a GPU starts.
- Summarise this lease in three lines.
- Two years, renews automatically, and either side can leave with 60 days’ notice.
- Draft a reply asking for 30 days.
What private means here.
-
No logs
Prompts and answers pass through memory only.
-
No training
Never, on anything you send.
-
History is yours
Chat history stays in your browser, encrypted, and only if you turn it on.
-
What we keep
Request metadata for billing and reliability: time, model, token counts and status. Nothing of the content.
Standard chat requests run on GPUs that also mine: our pool briefly checks a small slice of intermediate model values in memory, then discards them. Private requests are never mined. Compute privacy
Open models in the chat.
| Model |
|---|
| Llama 3.1 8B Instruct |
| Gemma 4 31B IT |
Private deployment for teams.
Dedicated GPUs, your own routing rules, no mining on your traffic. Talk to us.