…Download

Community performance summary

AMD Radeon Vega ×2

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

3 signed runs · 0 verified · 1 tester
Median prompt427.04tok/s · 319.02 to 793.59
Median generation23.94tok/s · 14.86 to 27.24
Configurations1Q4_K_M
Last observedJul 23GCN / 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 GPU3 groups
Qwen3 4B1 run→Qwen3 Coder 30B-A3B1 run→Gemma 4 12B1 run→
macOS

macOS 15.7.5 Sequoia

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 4BQ4_K_M · Dense · Signed ToshLLM app · amd-gpu · Metal · comparison
AMD Radeon Vega ×231.97 GB VRAM
GCN / Vega
793.59tok/s
27.24tok/s
ToshLLM communityJul 23, 2026
Qwen3 Coder 30B-A3BQ4_K_M · MoE · Signed ToshLLM app · amd-gpu · Metal · comparison
AMD Radeon Vega ×231.97 GB VRAM
GCN / Vega
427.04tok/s
23.94tok/s
ToshLLM communityJul 23, 2026
Gemma 4 12BQ4_K_M · Dense · Signed ToshLLM app · amd-gpu · Metal · comparison
AMD Radeon Vega ×231.97 GB VRAM
GCN / Vega
319.02tok/s
14.86tok/s
ToshLLM communityJul 23, 2026