🔌
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 iGPT-L?
Yes — here's how fast and up to what context, with real measured numbers.
iGPT-L on the Intel Gaudi 3
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 1732k | 787.8-978 t/s | ~8.5-10.6 (slow) |
| Q8 | yes | 1753k | 1575.6-1955.9 t/s | ~17-21.1 (slow) |
| Q4 | yes | 1762k | 2813.6-3492.8 t/s | ~30.4-37.8 (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 iGPT-L?▶
Yes. At Q4 it generates ~2813.6-3492.8 tok/s and fits up to 1762k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~30.4-37.8 tok/s.
How much context fits?▶
At Q4, up to about 1762k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.