🔌
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 UL2?
Yes — here's how fast and up to what context, with real measured numbers.
UL2 on the Intel Gaudi 3
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 291k | 53.7-66.6 t/s | ~0.6-0.7 (slow) |
| Q8 | yes | 367k | 107.3-133.2 t/s | ~1.2-1.4 (slow) |
| Q4 | yes | 401k | 191.6-237.9 t/s | ~2.1-2.6 (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 Intel Gaudi 3 run UL2?▶
Yes. At Q4 it generates ~191.6-237.9 tok/s and fits up to 401k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~2.1-2.6 tok/s.
How much context fits?▶
At Q4, up to about 401k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.