Community performance summary

AMD Radeon GFX10, 16 GB

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

9 signed runs · 0 verified · 1 tester
Median prompt1131.82tok/s · 720.61 to 1881.52
Median generation72.12tok/s · 45.53 to 105.77
Configurations5FP16 · Q4_0 · Q4_K_M
Last observedAug 6RDNA

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.

Recorded evidence

Inspect every run.

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

1 to 8 of 9 configurations

Verified benchmark Community submission
Model / quantHardwareArchitecturePromptGenerationSource
LLaMA V2 7BQ4_0 · Dense · Signed ToshLLM app · amd-gpu · Metal · comparison
RDNA
1508.5tok/s
105.77tok/s
ToshLLM communityAug 6, 2026
Gemma 4 19BQ4_K_M · MoE · Signed ToshLLM app · amd-gpu · Metal · comparison
RDNA
1881.52tok/s
99.11tok/s
ToshLLM communityAug 6, 2026
GLM 4.7 Flash 23B-A3BQ4_K_M · MoE · Signed ToshLLM app · ncmoe 1 · amd-gpu · Metal · comparison
RDNA
1341.71tok/s
94.83tok/s
ToshLLM communityAug 6, 2026
Gemma 4 A4B 98e V6 Coder ItQ4_K_M · MoE · Signed ToshLLM app · amd-gpu · Metal · comparison
RDNA
1773.77tok/s
94.17tok/s
ToshLLM communityAug 6, 2026
Models Zai Org GLM 4.6V FlashQ4_K_M · Dense · Signed ToshLLM app · amd-gpu · Metal · comparison
RDNA
1131.82tok/s
72.12tok/s
ToshLLM communityAug 6, 2026
Qwen3.6 12BQ5_K_M · Dense · Signed ToshLLM app · amd-gpu · Metal · comparison
RDNA
855.26tok/s
52.83tok/s
ToshLLM communityAug 6, 2026
Gemma4 V2 Merged 12BFP16 · Dense · Signed ToshLLM app · amd-gpu · Metal · comparison
RDNA
833tok/s
50.83tok/s
ToshLLM communityAug 6, 2026
Gemma 4 12BUD-Q4_K_XL · Dense · Signed ToshLLM app · amd-gpu · Metal · comparison
RDNA
825.73tok/s
48.46tok/s
ToshLLM communityJul 24, 2026
UncensoredLM DeepSeek R1 Qwen 13.5BQ5_K_M · Dense · Signed ToshLLM app · amd-gpu · Metal · comparison
RDNA
720.61tok/s
45.53tok/s
ToshLLM communityAug 6, 2026