New: connect Claude & other AIs to GenAIList over MCP — research the catalog and contribute to the shared knowledge base. Learn how →
Can the Instinct MI430X run HyenaDNA?
Yes — here's how fast and up to what context, with real measured numbers.
HyenaDNA on the Instinct MI430X
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 6041k | 1112045.5-1359166.7 t/s | ~1909.1-2333.3 (slow) |
| Q8 | yes | 6041k | 2224090.9-2718333.3 t/s | ~3818.2-4666.7 (slow) |
| Q4 | yes | 6042k | 3971590.9-4854166.7 t/s | ~6818.2-8333.3 (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 HyenaDNA?▶
Yes. At Q4 it generates ~3971590.9-4854166.7 tok/s and fits up to 6042k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~6818.2-8333.3 tok/s.
How much context fits?▶
At Q4, up to about 6042k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.