Qwen3 Coder 30B-A3B

Q4_K_M running on AMD Radeon RX 480 with macOS 15.8.1 Sequoia

PERFORMANCE / TOKENS PER SECONDGCN / Vega

Prompt processing102.36tok/s

#326 in index

Token generation3.41tok/s

#348 in index

Run fingerprint

Enough context to make the number useful.

ModelQwen3 Coder 30B-A3BMoE architecture

QuantizationQ4_K_MExact submitted variant

GPUAMD Radeon RX 4804 GB VRAM

GPU architectureGCN / VegaMetal backend

Test context512prompt tokens

Model familyMoESigned ToshLLM app · ncmoe 48 · amd-gpu · Metal

ToshLLM versionv0.87.22App 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.