🔌
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 Qwen2-VL-2B?
Yes — here's how fast and up to what context, with real measured numbers.
Qwen2-VL-2B on the Intel Gaudi 3
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 1713k | 536.5-666 t/s | ~5.8-7.2 (slow) |
| Q8 | yes | 1743k | 1073-1332 t/s | ~11.6-14.4 (slow) |
| Q4 | yes | 1757k | 1916.1-2378.6 t/s | ~20.7-25.7 (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 Qwen2-VL-2B?▶
Yes. At Q4 it generates ~1916.1-2378.6 tok/s and fits up to 1757k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~20.7-25.7 tok/s.
How much context fits?▶
At Q4, up to about 1757k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.