LOCAL AIregistry
‹ all GPUs

RTX 3060 Ti

NVIDIA · 8 GB · 448 GB/s
The recipe
Recommended
Qwen3.5-9B
EXL3 3 bpwTabbyAPI64K context

Best for Fast answers and light agent work on 8–12 GB cards.

86 tok/sdecode
1,258 tok/sprefill
64Kcontext
✓ Thinks✓ Tools✓ Long context
Tested on this card · Sep 25, 2026