Qwen3 4B

Q4_K_M running on RX 6900 XT

PERFORMANCE / TOKENS PER SECONDRDNA 2

Prompt processing291.4tok/s

#15 in index

Token generation97.5tok/s

#2 in index

Run fingerprint

Enough context to make the number useful.

ModelQwen3 4BDense architecture

QuantizationQ4_K_MExact submitted variant

GPURX 6900 XT16 GB VRAM

GPU architectureRDNA 2Metal backend

Test context512prompt tokens

Model familyDenseStandard configuration

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.