Someone from the community just pushed OrcaSAQ-2 Cyber 27B to 219K context on 2ร RTX 3060s.
68 tok/s. 15.7GB model. Correct long-context retrieval.
His verdict: โthe best local 27B result I have measured so far.โ
This is why we built SAQ-2: more intelligence per GB for everyone, on any hardware.