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 ESM2-650M?
Yes — here's how fast and up to what context, with real measured numbers.
ESM2-650M on the Instinct MI430X
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 6022k | 11291.5-13800.8 t/s | ~19.4-23.7 (slow) |
| Q8 | yes | 6032k | 22583.1-27601.5 t/s | ~38.8-47.4 (slow) |
| Q4 | yes | 6036k | 40326.9-49288.5 t/s | ~69.2-84.6 (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 ESM2-650M?▶
Yes. At Q4 it generates ~40326.9-49288.5 tok/s and fits up to 6036k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~69.2-84.6 tok/s.
How much context fits?▶
At Q4, up to about 6036k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.