🔌
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 AMD Instinct MI325X run P-LLama3?
Yes — here's how fast and up to what context, with real measured numbers.
P-LLama3 on the AMD Instinct MI325X
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 1663k | 236.3-288.8 t/s | ~1.6-1.9 (slow) |
| Q8 | yes | 1724k | 472.5-577.5 t/s | ~3.2-3.9 (slow) |
| Q4 | yes | 1751k | 843.8-1031.3 t/s | ~5.6-6.9 (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 AMD Instinct MI325X run P-LLama3?▶
Yes. At Q4 it generates ~843.8-1031.3 tok/s and fits up to 1751k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~5.6-6.9 tok/s.
How much context fits?▶
At Q4, up to about 1751k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.