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 OPT-2.7B (finetuned on WT2)?
Yes — here's how fast and up to what context, with real measured numbers.
OPT-2.7B (finetuned on WT2) on the Instinct MI430X
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 5959k | 2718.3-3322.4 t/s | ~4.7-5.7 (slow) |
| Q8 | yes | 6000k | 5436.7-6644.8 t/s | ~9.3-11.4 (slow) |
| Q4 | yes | 6018k | 9708.3-11865.7 t/s | ~16.7-20.4 (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 OPT-2.7B (finetuned on WT2)?▶
Yes. At Q4 it generates ~9708.3-11865.7 tok/s and fits up to 6018k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~16.7-20.4 tok/s.
How much context fits?▶
At Q4, up to about 6018k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.