🔌
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 NLLB?
Yes — here's how fast and up to what context, with real measured numbers.
NLLB on the Intel Gaudi 3
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 22k | 19.7-24.4 t/s | ~0.2-0.3 (slow) |
| Q8 | yes | 188k | 39.4-48.9 t/s | ~0.4-0.5 (slow) |
| Q4 | yes | 262k | 70.3-87.3 t/s | ~0.8-0.9 (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 Intel Gaudi 3 run NLLB?▶
Yes. At Q4 it generates ~70.3-87.3 tok/s and fits up to 262k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~0.8-0.9 tok/s.
How much context fits?▶
At Q4, up to about 262k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.