🔌
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 Minitron 8B?
Yes — here's how fast and up to what context, with real measured numbers.
Minitron 8B on the Intel Gaudi 3
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 760k | 129.3-160.5 t/s | ~1.4-1.7 (slow) |
| Q8 | yes | 824k | 258.6-321 t/s | ~2.8-3.5 (slow) |
| Q4 | yes | 852k | 461.7-573.1 t/s | ~5-6.2 (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 Minitron 8B?▶
Yes. At Q4 it generates ~461.7-573.1 tok/s and fits up to 852k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~5-6.2 tok/s.
How much context fits?▶
At Q4, up to about 852k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.