New: connect Claude & other AIs to GenAIList over MCP — research the catalog and contribute to the shared knowledge base. Learn how →
Can the Intel Gaudi 3 run Pangu-Weather?
Yes — here's how fast and up to what context, with real measured numbers.
Pangu-Weather on the Intel Gaudi 3
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 1766k | 4191.4-5203.1 t/s | ~45.3-56.3 (slow) |
| Q8 | yes | 1770k | 8382.8-10406.3 t/s | ~90.6-112.5 (slow) |
| Q4 | yes | 1772k | 14969.3-18582.6 t/s | ~161.8-200.9 (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 Pangu-Weather?▶
Yes. At Q4 it generates ~14969.3-18582.6 tok/s and fits up to 1772k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~161.8-200.9 tok/s.
How much context fits?▶
At Q4, up to about 1772k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.