Frontier AI on
metal we physically own.
Uncensored open-weight models. Zero server-side logs. Pay in crypto. Your prompts never transit a stranger's GPU — a claim rented-cloud AI structurally cannot make.
$ ask --private
you › write a raw socket port scanner in Rust
glm › Here's a complete async connect-scanner
async fn scan(ip: IpAddr, port: u16) …
nothing above was logged. this conversation lives only in your browser.
No logs. Nothing to leak.
Your prompts travel over TLS, get computed, and are never written to our disks — the gateway keeps only a token counter for billing. Your history lives only in your browser, so there's nothing on our side to log, leak, or subpoena.
GHOSTMETAL · SERVERS
Owned hardware
Inference runs on our own servers in a jurisdiction we choose. No third-party GPU ever sees your plaintext.
No logs, by design
The gateway keeps only a token counter for billing. Prompts and outputs are never written to disk.
Local-first history
Conversations live only in your browser. One button burns them. Nothing to subpoena from us.
Buy compute, not a subscription
Every plan is a pack of credits. Spend them, or stake and draw compute forever.
Free
free start · we log nothing
- › GLM-5.2 fast lane
- › 32K context
- › uncensored chat
- › local-only history
Plus
13,500 credits/mo
- › GLM-5.2 flagship
- › 128K context
- › FLUX images
- › voice · E2E sync
API
prepaid · BTC/XMR
- › OpenAI drop-in
- › $0.50/M output
- › no moderation
- › SillyTavern ready
A token backed by hardware we own.
Not vaporware yield — a claim on real inference. Stake $CMP, draw a daily compute allowance forever, keep the principal. Miners earn it by serving the network.