🔌
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 Multi-Token Prediction 7B?
Yes — here's how fast and up to what context, with real measured numbers.
Multi-Token Prediction 7B on the FuriosaAI RNGD
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 223k | 72.8-88.4 t/s | ~1.9-2.4 (slow) |
| Q8 | yes | 274k | 145.5-176.9 t/s | ~3.9-4.7 (slow) |
| Q4 | yes | 297k | 259.9-315.8 t/s | ~6.9-8.4 (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 Multi-Token Prediction 7B?▶
Yes. At Q4 it generates ~259.9-315.8 tok/s and fits up to 297k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~6.9-8.4 tok/s.
How much context fits?▶
At Q4, up to about 297k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.