Qwen K-quant

K-quant running on RX 580

PERFORMANCE / TOKENS PER SECONDGCN / Vega

Prompt processing285tok/s

#16 in index

Token generation51tok/s

#11 in index

Run fingerprint

Enough context to make the number useful.

ModelQwen K-quantDense architecture

QuantizationK-quantExact submitted variant

GPURX 5808 GB VRAM

GPU architectureGCN / VegaMetal backend

Test context512prompt tokens

Model familyDenseWave64 path; model details pending verification

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.