New: connect Claude & other AIs to GenAIList over MCP — research the catalog and contribute to the shared knowledge base. Learn how →
Can the FuriosaAI RNGD run AWD-LSTM + MoS + Partial Shuffled?
Yes — here's how fast and up to what context, with real measured numbers.
AWD-LSTM + MoS + Partial Shuffled on the FuriosaAI RNGD
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 650k | 13928.6-16928.6 t/s | ~371.4-451.4 (slow) |
| Q8 | yes | 650k | 27857.1-33857.1 t/s | ~742.9-902.9 (slow) |
| Q4 | yes | 651k | 49744.9-60459.2 t/s | ~1326.5-1612.2 (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 AWD-LSTM + MoS + Partial Shuffled?▶
Yes. At Q4 it generates ~49744.9-60459.2 tok/s and fits up to 651k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~1326.5-1612.2 tok/s.
How much context fits?▶
At Q4, up to about 651k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.