benchmark: AMD Ryzen 7 7700, Debian 12 Linux

[grigio]

• OS: Debian 12
• CPU: AMD Ryzen 7 7700
• RAM: 48GB

docker run --rm -v $PWD/models:/models -v $PWD/output/:/output sd -m /models/v1-5-pruned-emaonly-ggml-model-f16.bin -p "a lovely cat" -o /models/out.png

Details

Option:
n_threads: 8
mode: txt2img
model_path: /models/v1-5-pruned-emaonly-ggml-model-f16.bin
output_path: output.png
init_img:
prompt: a lovely cat
negative_prompt:
cfg_scale: 7.00
width: 512
height: 512
sample_method: eular a
sample_steps: 20
strength: 0.75
seed: 42
System Info:
BLAS = 0
SSE3 = 1
AVX = 1
AVX2 = 1
AVX512 = 0
AVX512_VBMI = 0
AVX512_VNNI = 0
FMA = 1
NEON = 0
ARM_FMA = 0
F16C = 1
FP16_VA = 0
WASM_SIMD = 0
VSX = 0
[INFO] stable-diffusion.cpp:2692 - loading model from '/models/v1-5-pruned-emaonly-ggml-model-f16.bin'
[DEBUG] stable-diffusion.cpp:2700 - verifying magic
[DEBUG] stable-diffusion.cpp:2711 - loading hparams
[INFO] stable-diffusion.cpp:2717 - ftype: f16
[DEBUG] stable-diffusion.cpp:2723 - loading vocab
[DEBUG] stable-diffusion.cpp:2751 - ggml tensor size = 288 bytes
[DEBUG] stable-diffusion.cpp:2756 - clip params ctx size = 236.24 MB
[DEBUG] stable-diffusion.cpp:2775 - unet params ctx size = 1641.37 MB
[DEBUG] stable-diffusion.cpp:2796 - vae params ctx size = 95.51 MB
[DEBUG] stable-diffusion.cpp:2817 - preparing memory for the weights
[DEBUG] stable-diffusion.cpp:2833 - loading weights
[DEBUG] stable-diffusion.cpp:2937 - model size = 1969.67MB
[INFO] stable-diffusion.cpp:2942 - total params size = 1969.98MB (clip 235.01MB, unet 1640.46MB, vae 94.51MB)
[INFO] stable-diffusion.cpp:2948 - loading model from '/models/v1-5-pruned-emaonly-ggml-model-f16.bin' completed, taking 1.79s
[DEBUG] stable-diffusion.cpp:771 - parse 'a lovely cat' to [['a lovely cat', 1], ]
[DEBUG] stable-diffusion.cpp:358 - split prompt "a lovely cat" to tokens ["a", "lovely", "cat", ]
[DEBUG] stable-diffusion.cpp:2985 - condition context need 1.65MB static memory, with work_size needing 0.45MB
[DEBUG] stable-diffusion.cpp:3010 - building condition graph completed: 633 nodes, 210 leafs
[DEBUG] stable-diffusion.cpp:3018 - computing condition graph completed, taking 0.06s
[INFO] stable-diffusion.cpp:3064 - condition graph use 239.59MB of memory: params 235.01MB, runtime 4.58MB (static 1.65MB, dynamic 2.93MB)
[DEBUG] stable-diffusion.cpp:3073 - 236544 bytes of dynamic memory has not been released yet
[DEBUG] stable-diffusion.cpp:771 - parse '' to [['', 1], ]
[DEBUG] stable-diffusion.cpp:358 - split prompt "" to tokens []
[DEBUG] stable-diffusion.cpp:2985 - condition context need 1.65MB static memory, with work_size needing 0.45MB
[DEBUG] stable-diffusion.cpp:3010 - building condition graph completed: 633 nodes, 210 leafs
[DEBUG] stable-diffusion.cpp:3018 - computing condition graph completed, taking 0.03s
[INFO] stable-diffusion.cpp:3064 - condition graph use 239.59MB of memory: params 235.01MB, runtime 4.58MB (static 1.65MB, dynamic 2.93MB)
[DEBUG] stable-diffusion.cpp:3073 - 236544 bytes of dynamic memory has not been released yet
[INFO] stable-diffusion.cpp:3569 - get_learned_condition completed, taking 0.09s
[INFO] stable-diffusion.cpp:3585 - start sampling
[DEBUG] stable-diffusion.cpp:3122 - diffusion context need 12.13MB static memory, with work_size needing 10.00MB
[INFO] stable-diffusion.cpp:3269 - step 1 sampling completed, taking 7.08s
[DEBUG] stable-diffusion.cpp:3270 - diffusion graph use 566.34MB runtime memory: static 12.13MB, dynamic 554.21MB
[DEBUG] stable-diffusion.cpp:3274 - 65536 bytes of dynamic memory has not been released yet
[INFO] stable-diffusion.cpp:3269 - step 2 sampling completed, taking 7.30s
[DEBUG] stable-diffusion.cpp:3270 - diffusion graph use 566.34MB runtime memory: static 12.13MB, dynamic 554.21MB
[DEBUG] stable-diffusion.cpp:3274 - 65536 bytes of dynamic memory has not been released yet
[INFO] stable-diffusion.cpp:3269 - step 3 sampling completed, taking 6.71s
[DEBUG] stable-diffusion.cpp:3270 - diffusion graph use 566.34MB runtime memory: static 12.13MB, dynamic 554.21MB
[DEBUG] stable-diffusion.cpp:3274 - 65536 bytes of dynamic memory has not been released yet
[INFO] stable-diffusion.cpp:3269 - step 4 sampling completed, taking 6.65s
[DEBUG] stable-diffusion.cpp:3270 - diffusion graph use 566.34MB runtime memory: static 12.13MB, dynamic 554.21MB
[DEBUG] stable-diffusion.cpp:3274 - 65536 bytes of dynamic memory has not been released yet
[INFO] stable-diffusion.cpp:3269 - step 5 sampling completed, taking 6.70s
[DEBUG] stable-diffusion.cpp:3270 - diffusion graph use 566.34MB runtime memory: static 12.13MB, dynamic 554.21MB
[DEBUG] stable-diffusion.cpp:3274 - 65536 bytes of dynamic memory has not been released yet
[INFO] stable-diffusion.cpp:3269 - step 6 sampling completed, taking 6.85s
[DEBUG] stable-diffusion.cpp:3270 - diffusion graph use 566.34MB runtime memory: static 12.13MB, dynamic 554.21MB
[DEBUG] stable-diffusion.cpp:3274 - 65536 bytes of dynamic memory has not been released yet
[INFO] stable-diffusion.cpp:3269 - step 7 sampling completed, taking 6.79s
[DEBUG] stable-diffusion.cpp:3270 - diffusion graph use 566.34MB runtime memory: static 12.13MB, dynamic 554.21MB
[DEBUG] stable-diffusion.cpp:3274 - 65536 bytes of dynamic memory has not been released yet
[INFO] stable-diffusion.cpp:3269 - step 8 sampling completed, taking 6.76s
[DEBUG] stable-diffusion.cpp:3270 - diffusion graph use 566.34MB runtime memory: static 12.13MB, dynamic 554.21MB
[DEBUG] stable-diffusion.cpp:3274 - 65536 bytes of dynamic memory has not been released yet
[INFO] stable-diffusion.cpp:3269 - step 9 sampling completed, taking 6.73s
[DEBUG] stable-diffusion.cpp:3270 - diffusion graph use 566.34MB runtime memory: static 12.13MB, dynamic 554.21MB
[DEBUG] stable-diffusion.cpp:3274 - 65536 bytes of dynamic memory has not been released yet
[INFO] stable-diffusion.cpp:3269 - step 10 sampling completed, taking 6.67s
[DEBUG] stable-diffusion.cpp:3270 - diffusion graph use 566.34MB runtime memory: static 12.13MB, dynamic 554.21MB
[DEBUG] stable-diffusion.cpp:3274 - 65536 bytes of dynamic memory has not been released yet
[INFO] stable-diffusion.cpp:3269 - step 11 sampling completed, taking 6.74s
[DEBUG] stable-diffusion.cpp:3270 - diffusion graph use 566.34MB runtime memory: static 12.13MB, dynamic 554.21MB
[DEBUG] stable-diffusion.cpp:3274 - 65536 bytes of dynamic memory has not been released yet
[INFO] stable-diffusion.cpp:3269 - step 12 sampling completed, taking 6.75s
[DEBUG] stable-diffusion.cpp:3270 - diffusion graph use 566.34MB runtime memory: static 12.13MB, dynamic 554.21MB
[DEBUG] stable-diffusion.cpp:3274 - 65536 bytes of dynamic memory has not been released yet
[INFO] stable-diffusion.cpp:3269 - step 13 sampling completed, taking 6.66s
[DEBUG] stable-diffusion.cpp:3270 - diffusion graph use 566.34MB runtime memory: static 12.13MB, dynamic 554.21MB
[DEBUG] stable-diffusion.cpp:3274 - 65536 bytes of dynamic memory has not been released yet
[INFO] stable-diffusion.cpp:3269 - step 14 sampling completed, taking 6.62s
[DEBUG] stable-diffusion.cpp:3270 - diffusion graph use 566.34MB runtime memory: static 12.13MB, dynamic 554.21MB
[DEBUG] stable-diffusion.cpp:3274 - 65536 bytes of dynamic memory has not been released yet
[INFO] stable-diffusion.cpp:3269 - step 15 sampling completed, taking 6.68s
[DEBUG] stable-diffusion.cpp:3270 - diffusion graph use 566.34MB runtime memory: static 12.13MB, dynamic 554.21MB
[DEBUG] stable-diffusion.cpp:3274 - 65536 bytes of dynamic memory has not been released yet
[INFO] stable-diffusion.cpp:3269 - step 16 sampling completed, taking 6.89s
[DEBUG] stable-diffusion.cpp:3270 - diffusion graph use 566.34MB runtime memory: static 12.13MB, dynamic 554.21MB
[DEBUG] stable-diffusion.cpp:3274 - 65536 bytes of dynamic memory has not been released yet
[INFO] stable-diffusion.cpp:3269 - step 17 sampling completed, taking 6.64s
[DEBUG] stable-diffusion.cpp:3270 - diffusion graph use 566.34MB runtime memory: static 12.13MB, dynamic 554.21MB
[DEBUG] stable-diffusion.cpp:3274 - 65536 bytes of dynamic memory has not been released yet
[INFO] stable-diffusion.cpp:3269 - step 18 sampling completed, taking 6.78s
[DEBUG] stable-diffusion.cpp:3270 - diffusion graph use 566.34MB runtime memory: static 12.13MB, dynamic 554.21MB
[DEBUG] stable-diffusion.cpp:3274 - 65536 bytes of dynamic memory has not been released yet
[INFO] stable-diffusion.cpp:3269 - step 19 sampling completed, taking 6.69s
[DEBUG] stable-diffusion.cpp:3270 - diffusion graph use 566.34MB runtime memory: static 12.13MB, dynamic 554.21MB
[DEBUG] stable-diffusion.cpp:3274 - 65536 bytes of dynamic memory has not been released yet
[INFO] stable-diffusion.cpp:3269 - step 20 sampling completed, taking 6.65s
[DEBUG] stable-diffusion.cpp:3270 - diffusion graph use 566.34MB runtime memory: static 12.13MB, dynamic 554.21MB
[DEBUG] stable-diffusion.cpp:3274 - 65536 bytes of dynamic memory has not been released yet
[INFO] stable-diffusion.cpp:3289 - diffusion graph use 2206.80MB of memory: params 1640.46MB, runtime 566.34MB (static 12.13MB, dynamic 554.21MB)
[DEBUG] stable-diffusion.cpp:3297 - 65536 bytes of dynamic memory has not been released yet
[INFO] stable-diffusion.cpp:3590 - sampling completed, taking 135.38s
[DEBUG] stable-diffusion.cpp:3460 - vae context need 1.14MB static memory, with work_size needing 0.00MB
[DEBUG] stable-diffusion.cpp:3490 - computing vae graph completed, taking 13.03s
[INFO] stable-diffusion.cpp:3506 - vae graph use 1759.65MB of memory: params 94.51MB, runtime 1665.14MB (static 1.14MB, dynamic 1664.00MB)
[DEBUG] stable-diffusion.cpp:3514 - 3145728 bytes of dynamic memory has not been released yet
[INFO] stable-diffusion.cpp:3603 - decode_first_stage completed, taking 13.18s
[INFO] stable-diffusion.cpp:3611 - txt2img completed in 148.65s, use 2301.31MB of memory: peak params memory 1969.98MB, peak runtime memory 1665.14MB
save result image to 'output.png'

[INFO]  stable-diffusion.cpp:3603 - decode_first_stage completed, taking 13.18s
[INFO]  stable-diffusion.cpp:3611 - txt2img completed in 148.65s, use 2301.31MB of memory: peak params memory 1969.98MB, peak runtime memory 1665.14MB
save result image to 'output.png'