🔌
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 Instinct MI430X run LUXIA-21.4B-Alignment?
Yes — here's how fast and up to what context, with real measured numbers.
LUXIA-21.4B-Alignment on the Instinct MI430X
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 1347k | 343-419.2 t/s | ~0.6-0.7 (slow) |
| Q8 | yes | 1429k | 685.9-838.4 t/s | ~1.2-1.4 (slow) |
| Q4 | yes | 1465k | 1224.9-1497.1 t/s | ~2.1-2.6 (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 Instinct MI430X run LUXIA-21.4B-Alignment?▶
Yes. At Q4 it generates ~1224.9-1497.1 tok/s and fits up to 1465k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~2.1-2.6 tok/s.
How much context fits?▶
At Q4, up to about 1465k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.