🔌
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 GMKtec EVO-X3 run ALBERT-xxlarge?
Yes — here's how fast and up to what context, with real measured numbers.
ALBERT-xxlarge on the GMKtec EVO-X3
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 1728k | 354-430.3 t/s | ~13.8-16.8 (slow) |
| Q8 | yes | 1731k | 708.1-860.6 t/s | ~27.7-33.6 (slow) |
| Q4 | yes | 1733k | 1264.4-1536.8 t/s | ~49.4-60 (slow) |
Estimate (memory-bound). Beyond "max context" the KV cache spills to SSD swap and speed drops to the "beyond" figure.
FAQ
Can the GMKtec EVO-X3 run ALBERT-xxlarge?▶
Yes. At Q4 it generates ~1264.4-1536.8 tok/s and fits up to 1733k context; beyond that it spills to SSD swap and slows to ~49.4-60 tok/s.
How much context fits?▶
At Q4, up to about 1733k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.