🔌
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 Jina Reranker v3.5?
Yes — here's how fast and up to what context, with real measured numbers.
Jina Reranker v3.5 on the GMKtec EVO-X3
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 1717k | 138.7-168.5 t/s | ~5.4-6.6 (slow) |
| Q8 | yes | 1726k | 277.3-337.1 t/s | ~10.8-13.2 (slow) |
| Q4 | yes | 1730k | 495.2-601.9 t/s | ~19.3-23.5 (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 Jina Reranker v3.5?▶
Yes. At Q4 it generates ~495.2-601.9 tok/s and fits up to 1730k context; beyond that it spills to SSD swap and slows to ~19.3-23.5 tok/s.
How much context fits?▶
At Q4, up to about 1730k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.