🔌
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 ERNIE-4.5-21B-A3B?
Yes — here's how fast and up to what context, with real measured numbers.
ERNIE-4.5-21B-A3B on the FuriosaAI RNGD
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 3k | 23.2-28.2 t/s | ~0.6-0.8 (slow) |
| Q8 | yes | 83k | 46.4-56.4 t/s | ~1.2-1.5 (slow) |
| Q4 | yes | 118k | 82.9-100.8 t/s | ~2.2-2.7 (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 ERNIE-4.5-21B-A3B?▶
Yes. At Q4 it generates ~82.9-100.8 tok/s and fits up to 118k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~2.2-2.7 tok/s.
How much context fits?▶
At Q4, up to about 118k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.