🔌
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 HiDream-I1?
Yes — here's how fast and up to what context, with real measured numbers.
HiDream-I1 on the Instinct MI430X
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 1373k | 407.8-498.4 t/s | ~0.7-0.9 (slow) |
| Q8 | yes | 1442k | 815.5-996.7 t/s | ~1.4-1.7 (slow) |
| Q4 | yes | 1472k | 1456.3-1779.9 t/s | ~2.5-3.1 (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 HiDream-I1?▶
Yes. At Q4 it generates ~1456.3-1779.9 tok/s and fits up to 1472k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~2.5-3.1 tok/s.
How much context fits?▶
At Q4, up to about 1472k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.