🔌
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 Unified-IO (XL)?
Yes — here's how fast and up to what context, with real measured numbers.
Unified-IO (XL) on the FuriosaAI RNGD
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 562k | 166.7-202.6 t/s | ~4.4-5.4 (slow) |
| Q8 | yes | 606k | 333.3-405.1 t/s | ~8.9-10.8 (slow) |
| Q4 | yes | 626k | 595.2-723.4 t/s | ~15.9-19.3 (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 Unified-IO (XL)?▶
Yes. At Q4 it generates ~595.2-723.4 tok/s and fits up to 626k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~15.9-19.3 tok/s.
How much context fits?▶
At Q4, up to about 626k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.