Qwen3 8B

Q4_K_M running on AMD Radeon RX 6600 XT with macOS 15.7.7 Sequoia

PERFORMANCE / TOKENS PER SECONDRDNA 2

Prompt processing570.26tok/s

#92 in index

Token generation42.02tok/s

#100 in index

Run fingerprint

Enough context to make the number useful.

ModelQwen3 8BDense architecture

QuantizationQ4_K_MExact submitted variant

GPUAMD Radeon RX 6600 XT7.98 GB VRAM

GPU architectureRDNA 2Metal backend

Test context512prompt tokens

Model familyDenseSigned ToshLLM app · amd-gpu · Metal

ToshLLM versionv0.83.7App 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.