Alpha · Umbra is in testing. No payments or payouts yet; all credit is free while we tune the network.

Playground

try any model on attested Apple hardware · nothing leaves in a log
Verify attestation
Live playgroundready to route through the coordinator
prompts are not stored
loading
Fetching live model catalog…
fetching catalog…
Trust level
SDK envelope
⌘⏎ to send
Response
Send a prompt to begin.Streams token-by-token, then shows usage, latency, route, and model.
usage, latency, and routing appear after a run

How Umbra sees your SDK call same wire · two envelopes

Both SDKs ride the same internal core: trust-level routing, prompt-hop crypto, and the token meter. Toggle the SDK above and these examples re-render with your model + settings.

OpenAI SDK
active

Standard /v1/chat/completions surface. Umbra-specific extension is extra_body.trust_level.

Anthropic SDK
switch →

/v1/messages surface. Same key. Anthropic-specific: max_tokens is required and anthropic-version is read on the wire.

Recent playground runs view usage →

WhenModelTrustTokensRoutingCost
No playground runs yet — send your first prompt above.

Continuerelated in the console

Browse the catalogswitch the modelSwitch to Umbraone base_url swap in your codeView usage & billingtoken cost so farVerify what ranread a response receipt