Community performance summary

AMD Radeon Pro Vega 64

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

2 signed runs · 0 verified · 1 tester
Median prompt338.58tok/s · 104.51 to 572.65
Median generation38.31tok/s · 13.11 to 63.51
Configurations2MXFP4 · UD-IQ2_S
Last observedSep 24GCN / 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
GPT OSS 20B1 run→Qwen3.8 27B1 run→
macOS

macOS 14.8.7 Sonoma

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
GPT OSS 20BMXFP4 · MoE · Signed ToshLLM app · amd-gpu · Metal · comparison
GCN / Vega
572.65tok/s
63.51tok/s
goroSep 24, 2026
Qwen3.8 27BUD-IQ2_S · Dense · Signed ToshLLM app · amd-gpu · Metal · comparison
GCN / Vega
104.51tok/s
13.11tok/s
goroSep 24, 2026