processors · furiosa

FuriosaAI RNGD

Tensor-contraction-processor inference card aimed at LLM serving at low power — 150 W with 48 GB HBM3, targeting perf-per-watt rather than peak throughput.

Verify

What this registry has checked

Closed. A third party reports this model and we have recorded their numbers. Nothing of our own: no weights seen, no publisher hash held. Treat every figure here as their claim.

Method	vendor-published-specification

Re-check this yourself → Source → Full record →

What is recorded

Publisher	furiosa
Origin	South Korea
Licence	vendor

Elsewhere in the registry

All processors → Browse with filters → What will run on my hardware? →

Registry record as of 2026-08-12. Identity, licence and parameter count are properties of a release and do not move; tier can, and the verify link above re-checks it against the live registry rather than this page.