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 SRU++ Large?
Yes — here's how fast and up to what context, with real measured numbers.
SRU++ Large on the Instinct MI430X
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 6034k | 31365.4-38335.5 t/s | ~53.8-65.8 (slow) |
| Q8 | yes | 6038k | 62730.8-76670.9 t/s | ~107.7-131.6 (slow) |
| Q4 | yes | 6040k | 112019.2-136912.4 t/s | ~192.3-235 (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 SRU++ Large?▶
Yes. At Q4 it generates ~112019.2-136912.4 tok/s and fits up to 6040k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~192.3-235 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.