Appearance
Explore LLM and image benchmark results for Beelink GTR9 Pro (Ryzen AI Max+ 395), with hardware specifications and the software settings used in each test.
| Model | Benchmark | Configuration | GTR9 Pro | Olares One |
|---|---|---|---|---|
| Qwen3.8-27B | IFBench Evaluation time | llama.cpp Vulkan, UD-Q4_K_XL | 33.1 h | 8.7 h |
| Qwen3.8-27B | Inference speed Token generation, single request | llama.cpp Vulkan, UD-Q4_K_XL | 28 tok/s | 93 tok/s |
| Qwen-Image-2.1 Base | Text to image Per image | ComfyUI, int8 | 1.6 min | 12 s |
| Qwen-Image-2.1 Turbo | Image edit Per image | ComfyUI, int8 | 47 s | 5.2 s |
llama.cpp Vulkan, UD-Q4_K_XL
--mmproj mmproj-F16.gguf -c 104448 -ngl all -fa on -ctk q8_0 -ctv q8_0 --jinja -np 1 --spec-type draft-mtp --spec-draft-n-max 5llama.cpp Vulkan, UD-Q4_K_XL
--mmproj mmproj-F16.gguf -c 104448 -ngl all -fa on -ctk q8_0 -ctv q8_0 --jinja -np 4 -kvu --spec-type draft-mtp --spec-draft-n-max 5Qwen-Image-2.1 Base: Text to image
Qwen-Image-2.1 Turbo: Image edit
--listen 127.0.0.1 --port 8288 --extra-model-paths-config extra_model_paths.yaml --output-directory images --input-directory inputs --disable-dynamic-vram, COMFYUI_ENABLE_MIOPEN=1Qwen-Image-2.1 Base: Text to image
Qwen-Image-2.1 Turbo: Image edit
--listen 127.0.0.1 --port 8288 --extra-model-paths-config extra_model_paths.yaml --output-directory images --input-directory inputs --disable-dynamic-vram, COMFYUI_ENABLE_MIOPEN=1