🔌
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 ruDalle: Kandinsky 3.0?
Yes — here's how fast and up to what context, with real measured numbers.
ruDalle: Kandinsky 3.0 on the Instinct MI430X
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 2271k | 616.8-753.8 t/s | ~1.1-1.3 (slow) |
| Q8 | yes | 2344k | 1233.5-1507.6 t/s | ~2.1-2.6 (slow) |
| Q4 | yes | 2376k | 2202.7-2692.2 t/s | ~3.8-4.6 (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 ruDalle: Kandinsky 3.0?▶
Yes. At Q4 it generates ~2202.7-2692.2 tok/s and fits up to 2376k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~3.8-4.6 tok/s.
How much context fits?▶
At Q4, up to about 2376k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.