Gemma 4 26B A4B

Q4_K_M running on Radeon Pro Vega II

PERFORMANCE / TOKENS PER SECONDGCN / Vega

Prompt processing190.3tok/s

#19 in index

Token generation47.9tok/s

#12 in index

Run fingerprint

Enough context to make the number useful.

ModelGemma 4 26B A4BMoE architecture

QuantizationQ4_K_MExact submitted variant

GPURadeon Pro Vega II32 GB VRAM

GPU architectureGCN / VegaMetal backend

Test context512prompt tokens

Model familyMoEStandard configuration

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 result remains explicitly labeled as a community submission.