community-verified · measured pool data

Can the RTX 3060 12GB run Qwen2.5-0.5B-Instruct?

Yes — community-measured at 396.2 tok/s (2 runs at 8-bit).

Verified single-stream runs on the RTX 3060 12GB from the localmaxxing public pool. Every number on this page is measured — nothing here is estimated.

Quantization	Median tok/s	TTFT (ms)	Peak VRAM	Max context tested	Engines
8-bit	396.2	201	2.4 GB	4,608	llama.cpp

Go deeper

I have a goal — find models that fit my use case → I have hardware — browse everything my rig can run → All machine × model answer pages →