Community performance summary

AMD Radeon Pro Vega II Duo

2 local models measured on this GPU configuration. Compare quantizations, configurations and the full observed performance range.

2 signed runs · 0 verified · 2 testers
Median prompt223.77tok/s · 126.88 to 320.65
Median generation26.21tok/s · 10.48 to 41.94
Configurations1UD-Q8_K_XL
Last observedAug 25GCN / Vega

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.

Models on this GPU2 groups
Qwen3.8 27B1 runQwen3.6 35B-A3B1 run
macOS

macOS 26.5 Tahoe · macOS 26.5.2 Tahoe

Recorded evidence

Inspect every run.

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

1 to 8 of 2 configurations

Verified benchmark Community submission
Model / quantHardwareArchitecturePromptGenerationSource
Qwen3.6 35B-A3BUD-Q8_K_XL · MoE · Signed ToshLLM app · amd-gpu · Metal · comparison
GCN / Vega
320.65tok/s
41.94tok/s
ToshLLM communityAug 5, 2026
Qwen3.8 27BUD-Q8_K_XL · Dense · Signed ToshLLM app · amd-gpu · Metal · comparison
GCN / Vega
126.88tok/s
10.48tok/s
sedtamensum redditAug 25, 2026