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 Pythia-160m?
Yes — here's how fast and up to what context, with real measured numbers.
Pythia-160m on the FuriosaAI RNGD
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 646k | 3046.9-3703.1 t/s | ~81.3-98.8 (slow) |
| Q8 | yes | 648k | 6093.8-7406.3 t/s | ~162.5-197.5 (slow) |
| Q4 | yes | 650k | 10881.7-13225.4 t/s | ~290.2-352.7 (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 Pythia-160m?▶
Yes. At Q4 it generates ~10881.7-13225.4 tok/s and fits up to 650k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~290.2-352.7 tok/s.
How much context fits?▶
At Q4, up to about 650k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.