Qwen3.6 35B-A3B

UD-Q4_K_S running on AMD Radeon RX 6700 XT with macOS 26.6.2 Tahoe

PERFORMANCE / TOKENS PER SECONDRDNA 2

Prompt processing473.74tok/s

#131 in index

Token generation27.34tok/s

#179 in index

Run fingerprint

Enough context to make the number useful.

ModelQwen3.6 35B-A3BMoE architecture

QuantizationUD-Q4_K_SExact submitted variant

GPUAMD Radeon RX 6700 XT11.98 GB VRAM

GPU architectureRDNA 2Metal backend

Test context512prompt tokens

Model familyMoESigned ToshLLM app · ncmoe 24 · amd-gpu · Metal

ToshLLM versionv0.87.1App 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 run was reproduced by ToshLLM Lab.