🔌
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 EXAONE Deep 7.8B?
Yes — here's how fast and up to what context, with real measured numbers.
EXAONE Deep 7.8B on the Intel Gaudi 3
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 768k | 137.6-170.8 t/s | ~1.5-1.8 (slow) |
| Q8 | yes | 827k | 275.1-341.5 t/s | ~3-3.7 (slow) |
| Q4 | yes | 854k | 491.3-609.9 t/s | ~5.3-6.6 (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 EXAONE Deep 7.8B?▶
Yes. At Q4 it generates ~491.3-609.9 tok/s and fits up to 854k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~5.3-6.6 tok/s.
How much context fits?▶
At Q4, up to about 854k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.