Community performance summary

Qwen3 8B × Vega ×4 + W6800X Duo ×3

3 signed observations in one performance comparison. Open any run below to inspect its exact configuration and provenance.

3 signed runs · 0 verified · 2 testers
Median prompt266.06tok/s · 262.11 to 396.42
Median generation29.41tok/s · 17.59 to 33.27
Configurations1Q4_K_M
Last observedJul 20GCN / Vega + RDNA 2

What the results show

One clear view.
Every recorded run.

Measurements from comparable hardware and models are presented together. Medians show typical performance, while every signed run remains available below for inspection and sharing.

Quantizations observed1 groups
Q4_K_M3 runs
macOS

macOS 26.5 Tahoe

Recorded evidence

Inspect every run.

Open any result to review its exact hardware, model configuration, measurements and submission details.

1 to 8 of 3 configurations

Verified benchmark Community submission
Model / quantHardwareArchitecturePromptGenerationSource
Qwen3 8BQ4_K_M · Dense · Signed ToshLLM app · amd-gpu · Metal · comparison
GCN / Vega + RDNA 2
266.06tok/s
33.27tok/s
ToshLLM communityJul 20, 2026
Qwen3 8BQ4_K_M · Dense · Signed ToshLLM app · amd-gpu · Metal · comparison
GCN / Vega + RDNA 2
396.42tok/s
29.41tok/s
Flint IronstagJul 20, 2026
Qwen3 8BQ4_K_M · Dense · Signed ToshLLM app · amd-gpu · Metal · comparison
GCN / Vega + RDNA 2
262.11tok/s
17.59tok/s
Flint IronstagJul 20, 2026