🔌
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 Intel Gaudi 3 run XLMR-XXL?
Yes — here's how fast and up to what context, with real measured numbers.
XLMR-XXL on the Intel Gaudi 3
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 579k | 100.3-124.5 t/s | ~1.1-1.3 (slow) |
| Q8 | yes | 644k | 200.6-249 t/s | ~2.2-2.7 (slow) |
| Q4 | yes | 673k | 358.1-444.6 t/s | ~3.9-4.8 (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 Intel Gaudi 3 run XLMR-XXL?▶
Yes. At Q4 it generates ~358.1-444.6 tok/s and fits up to 673k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~3.9-4.8 tok/s.
How much context fits?▶
At Q4, up to about 673k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.