🔌
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 ProGen2-base?
Yes — here's how fast and up to what context, with real measured numbers.
ProGen2-base on the FuriosaAI RNGD
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 628k | 638.1-775.5 t/s | ~17-20.7 (slow) |
| Q8 | yes | 639k | 1276.2-1551 t/s | ~34-41.4 (slow) |
| Q4 | yes | 644k | 2278.9-2769.7 t/s | ~60.8-73.9 (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 ProGen2-base?▶
Yes. At Q4 it generates ~2278.9-2769.7 tok/s and fits up to 644k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~60.8-73.9 tok/s.
How much context fits?▶
At Q4, up to about 644k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.