🔌
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 Llama Nemotron Super v1.5?
Yes — here's how fast and up to what context, with real measured numbers.
Llama Nemotron Super v1.5 on the Intel Gaudi 3
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 56k | 21.9-27.2 t/s | ~0.2-0.3 (slow) |
| Q8 | yes | 205k | 43.8-54.4 t/s | ~0.5-0.6 (slow) |
| Q4 | yes | 271k | 78.2-97.1 t/s | ~0.8-1 (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 Llama Nemotron Super v1.5?▶
Yes. At Q4 it generates ~78.2-97.1 tok/s and fits up to 271k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~0.8-1 tok/s.
How much context fits?▶
At Q4, up to about 271k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.