🔌
New: connect Claude & other AIs to GenAIList over MCP — research the catalog and contribute to the shared knowledge base. Learn how →
// RUN — FITS
Can the FuriosaAI RNGD run ALBERT-xxlarge?
Yes — here's how fast and up to what context, with real measured numbers.
ALBERT-xxlarge on the FuriosaAI RNGD
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 644k | 2074.5-2521.3 t/s | ~55.3-67.2 (slow) |
| Q8 | yes | 647k | 4148.9-5042.6 t/s | ~110.6-134.5 (slow) |
| Q4 | yes | 649k | 7408.8-9004.6 t/s | ~197.6-240.1 (slow) |
Estimate (memory-bound). Beyond "max context" the KV cache spills to system RAM offload (depends on your system RAM) and speed drops to the "beyond" figure.
FAQ
Can the FuriosaAI RNGD run ALBERT-xxlarge?▶
Yes. At Q4 it generates ~7408.8-9004.6 tok/s and fits up to 649k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~197.6-240.1 tok/s.
How much context fits?▶
At Q4, up to about 649k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.