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 FvFold?
Yes — here's how fast and up to what context, with real measured numbers.
FvFold on the Intel Gaudi 3
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 1774k | 116506.5-144628.7 t/s | ~1259.5-1563.6 (slow) |
| Q8 | yes | 1774k | 233013-289257.5 t/s | ~2519.1-3127.1 (slow) |
| Q4 | yes | 1774k | 416094.6-516531.2 t/s | ~4498.3-5584.1 (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 FvFold?▶
Yes. At Q4 it generates ~416094.6-516531.2 tok/s and fits up to 1774k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~4498.3-5584.1 tok/s.
How much context fits?▶
At Q4, up to about 1774k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.