Gemma 4 12B

UD-Q4_K_XL running on AMD Radeon RX 6700 XT with macOS 26.5.1 Tahoe

PERFORMANCE / TOKENS PER SECONDRDNA 2

Prompt processing337.18tok/s

#64 in index

Token generation34.66tok/s

#57 in index

Run fingerprint

Enough context to make the number useful.

ModelGemma 4 12BDense architecture

QuantizationUD-Q4_K_XLExact submitted variant

GPUAMD Radeon RX 6700 XT11.98 GB VRAM

GPU architectureRDNA 2Metal backend

Test context512prompt tokens

Model familyDenseSigned ToshLLM app · amd-gpu · Metal

Prompt processing and generation are reported separately so memory bandwidth and decode behavior stay visible.

Quantization, GPU family, VRAM and context travel with the result instead of being hidden behind one score.

This run was reproduced by ToshLLM Lab.