GLM 4.7 Flash 23B-A3B

Q4_K_M running on AMD Radeon GFX10, 16 GB + AMD Radeon RX 5600 XT with macOS 15.7.8 Sequoia

PERFORMANCE / TOKENS PER SECONDRDNA + RDNA 1

Prompt processing925.86tok/s

#79 in index

Token generation32.66tok/s

#188 in index

Run fingerprint

Enough context to make the number useful.

ModelGLM 4.7 Flash 23B-A3BMoE architecture

QuantizationQ4_K_MExact submitted variant

GPUAMD Radeon GFX10, 16 GB + AMD Radeon RX 5600 XT21.97 GB VRAM

GPU architectureRDNA + RDNA 1Metal backend

Test context512prompt tokens

Model familyMoESigned ToshLLM app · amd-gpu · Metal

ToshLLM versionv0.83.16App 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.