Gemma4 26B-A4B

Q4_K_M running on AMD Radeon Pro 5600M with macOS 15.7.9 Sequoia

PERFORMANCE / TOKENS PER SECONDRDNA 1

Prompt processing152.43tok/s

#228 in index

Token generation14.95tok/s

#225 in index

Run fingerprint

Enough context to make the number useful.

ModelGemma4 26B-A4BMoE architecture

QuantizationQ4_K_MExact submitted variant

GPUAMD Radeon Pro 5600M7.98 GB VRAM

GPU architectureRDNA 1Metal backend

Test context512prompt tokens

Model familyMoESigned ToshLLM app · ncmoe 20 · amd-gpu · Metal

ToshLLM versionv0.87.0App version used to capture and sign this run

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.