🔌
New: connect Claude & other AIs to GenAIList over MCP — research the catalog and contribute to the shared knowledge base. Learn how →
// RUN — FITS
Can the Instinct MI440X run PLATO-XL?
Yes — here's how fast and up to what context, with real measured numbers.
PLATO-XL on the Instinct MI440X
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 2282k | 561.3-686 t/s | ~1.1-1.4 (slow) |
| Q8 | yes | 2349k | 1122.5-1372 t/s | ~2.3-2.8 (slow) |
| Q4 | yes | 2379k | 2004.5-2450 t/s | ~4.1-5 (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 MI440X run PLATO-XL?▶
Yes. At Q4 it generates ~2004.5-2450 tok/s and fits up to 2379k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~4.1-5 tok/s.
How much context fits?▶
At Q4, up to about 2379k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.