community-verified · measured pool data

Can the Radeon AI Pro R9700 32GB ×3 run gemma-4-12B-it?

Yes — community-measured at 19.3 tok/s (2 runs at 16-bit).

Verified single-stream runs on the Radeon AI Pro R9700 32GB ×3 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
16-bit	19.3	1447.6	—	794	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 →