🔌
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 phi-3.5-mini?
Yes — here's how fast and up to what context, with real measured numbers.
phi-3.5-mini on the FuriosaAI RNGD
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 535k | 128.3-155.9 t/s | ~3.4-4.2 (slow) |
| Q8 | yes | 593k | 256.6-311.8 t/s | ~6.8-8.3 (slow) |
| Q4 | yes | 618k | 458.2-556.9 t/s | ~12.2-14.8 (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 phi-3.5-mini?▶
Yes. At Q4 it generates ~458.2-556.9 tok/s and fits up to 618k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~12.2-14.8 tok/s.
How much context fits?▶
At Q4, up to about 618k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.