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 (ensemble): AWD-LSTM-DOC (fin) × 5 (WT2)?
Yes — here's how fast and up to what context, with real measured numbers.
(ensemble): AWD-LSTM-DOC (fin) × 5 (WT2) on the Instinct MI430X
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 6036k | 39673-48489.2 t/s | ~68.1-83.2 (slow) |
| Q8 | yes | 6039k | 79345.9-96978.4 t/s | ~136.2-166.5 (slow) |
| Q4 | yes | 6040k | 141689.2-173175.7 t/s | ~243.2-297.3 (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 (ensemble): AWD-LSTM-DOC (fin) × 5 (WT2)?▶
Yes. At Q4 it generates ~141689.2-173175.7 tok/s and fits up to 6040k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~243.2-297.3 tok/s.
How much context fits?▶
At Q4, up to about 6040k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.