🔌
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 Hymba?
Yes — here's how fast and up to what context, with real measured numbers.
Hymba on the Instinct MI430X
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 5996k | 4893-5980.3 t/s | ~8.4-10.3 (slow) |
| Q8 | yes | 6019k | 9786-11960.7 t/s | ~16.8-20.5 (slow) |
| Q4 | yes | 6029k | 17475-21358.3 t/s | ~30-36.7 (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 Instinct MI430X run Hymba?▶
Yes. At Q4 it generates ~17475-21358.3 tok/s and fits up to 6029k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~30-36.7 tok/s.
How much context fits?▶
At Q4, up to about 6029k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.