Hi
Here I am testing different combinations of samplers with some basic checkpoints. [Also each 4 pics of each checkpoint]
dynavisionXLAllInOneStylized_beta0371Bakedvae - [Image Post]
nightvisionXLPhotorealisticPortrait_alpha0650Bakedvae - [Image Post]
and the samplers I have used are
DPM++ 2M Karras
DPM++ SDE Karras
DPM++ 2M SDE Exponential
DPM++ 2M SDE Karras.
What I want to try is that if we try a little complex prompts, without using any LORA or EMBEDDINGS, what will be the output by each checkpoint, with each sampler.
I started the X/Y/Z plot first with all checkpoints and samplers, but then I saw that the output grid was too big to realise changes without zoom in the grid image.
So then I started with all checkpoints everytime, but limited samplers, so in the end I came up with 4 sampler at a time, which will be easy for everyone to see the difference, as more tahtn 4 also confuses brain to memorise difference.
So basically her is the prompt I entered:
Prompt::(boy and girl) north Indian wearing traditional dress and ornaments, standing surrounded water source, (girl wearing strip blouse with deep cleavage, big breasts), (boy wearing traditional dhoti and topless), wet (body, clothes), asian skin tone, realistic, finely detailed, quality,
Negative Prompt::nude, nsfw, (exposed breasts,exposed boobs:1.1), exposed vagina, exposed penis, exposed private parts, dark skin, sexual, (low resolution:1.3), (blurry:1.3), (worst quality:1.3), (low quality:1.3), collage, bad proportions, (watermark:1.3), letter,
Steps: 25, CFG scale: 3, Seed: 2554746932, Size: 1024x1536, Clip skip: 2
Script: X/Y/Z plot
X Type: Sampler, X Values: "DPM++ 2M Karras,DPM++ SDE Karras,DPM++ 2M SDE Exponential,DPM++ 2M SDE Karras"
Y Type: Checkpoint name, Y Values: "cherryPickerXL_v20.safetensors [e250118276],dreamshaperXL10_alpha2Xl10.safetensors [0f1b80cfe8],dynavisionXLAllInOneStylized_beta0371Bakedvae.safetensors [5394949ab6],miamodelSFWNSFWSDXL_v30.safetensors [71a9263b31],nightvisionXLPhotorealisticPortrait_alpha0650Bakedvae.safetensors [b88d82a292],protovisionXLHighFidelity3D_beta0520Bakedvae.safetensors [0538f9319d],sdvn6Realxl_detailface.safetensors [777f31751a],sdXL_v10VAEFix.safetensors [e6bb9ea85b],talmendoxlSDXL_v11Beta.safetensors [7fb8947de2],wyvernmix15XL_xlV11.safetensors [1ba2448476],xxmix9realisticsdxl_testV20.safetensors [ecb947b4ea]",
Automatic 1111 Version: v1.6.0
So above is the output.
Total time took: 4 hours 25 minutes 43 seconds.
Please see yourself what you think, and please also comment and provide feedback this is my first article, also I will be continuing the article for next samplers one-by-one.
You can also comment if you want me to check anything or try anything.
I am also providing output of console, so you can also check the total time taken for each step.
X/Y/Z plot will create 44 images on 1 4x11 grid. (Total steps to process: 1100)
100%|██████████████████████████| 25/25 [02:39<00:00, 6.37s/it]
100%|██████████████████████████| 25/25 [05:15<00:00, 12.62s/it]
100%|██████████████████████████| 25/25 [02:44<00:00, 6.60s/it]
100%|██████████████████████████| 25/25 [02:53<00:00, 6.94s/it]
Reusing loaded model cherryPickerXL_v20.safetensors [e250118276] to load dreamshaperXL10_alpha2Xl10.safetensors [0f1b80cfe8] | 100/1100 [13:51<1:51:55, 6.72s/it]
Loading weights [0f1b80cfe8] from /Users/ashutoshtyagi/Documents/Personal/gan/webui-sdxl/models/Stable-diffusion/dreamshaperXL10_alpha2Xl10.safetensors
Applying attention optimization: sub-quadratic... done.
Weights loaded in 26.4s (send model to cpu: 7.7s, load weights from disk: 0.8s, apply weights to model: 13.8s, move model to device: 4.1s).
100%|██████████████████████████| 25/25 [02:51<00:00, 6.86s/it]
100%|██████████████████████████| 25/25 [05:24<00:00, 12.98s/it]
100%|██████████████████████████| 25/25 [02:40<00:00, 6.41s/it]
100%|██████████████████████████| 25/25 [02:40<00:00, 6.42s/it]
Reusing loaded model dreamshaperXL10_alpha2Xl10.safetensors [0f1b80cfe8] to load dynavisionXLAllInOneStylized_beta0371Bakedvae.safetensors [5394949ab6] | 200/1100 [28:33<1:35:49, 6.39s/it]
Loading weights [5394949ab6] from /Users/ashutoshtyagi/Documents/Personal/gan/webui-sdxl/models/Stable-diffusion/dynavisionXLAllInOneStylized_beta0371Bakedvae.safetensors
Applying attention optimization: sub-quadratic... done.
Weights loaded in 28.0s (send model to cpu: 9.2s, load weights from disk: 0.8s, apply weights to model: 13.6s, move model to device: 4.4s).
100%|██████████████████████████| 25/25 [02:48<00:00, 6.74s/it]
100%|██████████████████████████| 25/25 [05:26<00:00, 13.04s/it]
100%|██████████████████████████| 25/25 [02:47<00:00, 6.70s/it]
100%|██████████████████████████| 25/25 [02:52<00:00, 6.91s/it]
Reusing loaded model dynavisionXLAllInOneStylized_beta0371Bakedvae.safetensors [5394949ab6] to load miamodelSFWNSFWSDXL_v30.safetensors [71a9263b31] | 300/1100 [43:47<1:25:20, 6.40s/it]
Loading weights [71a9263b31] from /Users/ashutoshtyagi/Documents/Personal/gan/webui-sdxl/models/Stable-diffusion/miamodelSFWNSFWSDXL_v30.safetensors
Applying attention optimization: sub-quadratic... done.
Weights loaded in 30.0s (send model to cpu: 10.6s, load weights from disk: 0.7s, apply weights to model: 14.2s, move model to device: 4.4s).
100%|██████████████████████████| 25/25 [02:46<00:00, 6.66s/it]
100%|██████████████████████████| 25/25 [05:18<00:00, 12.76s/it]
100%|██████████████████████████| 25/25 [02:40<00:00, 6.41s/it]
100%|██████████████████████████| 25/25 [02:40<00:00, 6.42s/it]
Reusing loaded model miamodelSFWNSFWSDXL_v30.safetensors [71a9263b31] to load nightvisionXLPhotorealisticPortrait_alpha0650Bakedvae.safetensors [b88d82a292] | 400/1100 [58:24<1:14:40, 6.40s/it]
Loading weights [b88d82a292] from /Users/ashutoshtyagi/Documents/Personal/gan/webui-sdxl/models/Stable-diffusion/nightvisionXLPhotorealisticPortrait_alpha0650Bakedvae.safetensors
Applying attention optimization: sub-quadratic... done.
Weights loaded in 27.9s (send model to cpu: 10.6s, load weights from disk: 0.7s, apply weights to model: 12.2s, move model to device: 4.4s).
100%|██████████████████████████| 25/25 [02:49<00:00, 6.79s/it]
100%|██████████████████████████| 25/25 [05:26<00:00, 13.07s/it]
100%|██████████████████████████| 25/25 [02:45<00:00, 6.60s/it]
100%|██████████████████████████| 25/25 [02:51<00:00, 6.85s/it]
Reusing loaded model nightvisionXLPhotorealisticPortrait_alpha0650Bakedvae.safetensors [b88d82a292] to load protovisionXLHighFidelity3D_beta0520Bakedvae.safetensors [0538f9319d] | 500/1100 [1:13:39<1:12:35, 7.26s/it]
Loading weights [0538f9319d] from /Users/ashutoshtyagi/Documents/Personal/gan/webui-sdxl/models/Stable-diffusion/protovisionXLHighFidelity3D_beta0520Bakedvae.safetensors
Applying attention optimization: sub-quadratic... done.
Weights loaded in 28.0s (send model to cpu: 9.2s, load weights from disk: 0.7s, apply weights to model: 13.6s, move model to device: 4.4s).
100%|██████████████████████████| 25/25 [02:47<00:00, 6.72s/it]
100%|██████████████████████████| 25/25 [05:40<00:00, 13.60s/it]
100%|██████████████████████████| 25/25 [02:43<00:00, 6.54s/it]
100%|██████████████████████████| 25/25 [02:52<00:00, 6.92s/it]
Reusing loaded model protovisionXLHighFidelity3D_beta0520Bakedvae.safetensors [0538f9319d] to load sdvn6Realxl_detailface.safetensors [777f31751a] | 600/1100 [1:29:39<56:48, 6.82s/it]
Loading weights [777f31751a] from /Users/ashutoshtyagi/Documents/Personal/gan/webui-sdxl/models/Stable-diffusion/sdvn6Realxl_detailface.safetensors
Applying attention optimization: sub-quadratic... done.
Weights loaded in 27.6s (send model to cpu: 9.3s, load weights from disk: 0.8s, apply weights to model: 13.2s, move model to device: 4.2s).
100%|██████████████████████████| 25/25 [02:53<00:00, 6.92s/it]
100%|██████████████████████████| 25/25 [05:34<00:00, 13.37s/it]
100%|██████████████████████████| 25/25 [02:50<00:00, 6.83s/it]
100%|██████████████████████████| 25/25 [02:49<00:00, 6.79s/it]
Reusing loaded model sdvn6Realxl_detailface.safetensors [777f31751a] to load sdXL_v10VAEFix.safetensors [e6bb9ea85b]████████████▌ | 700/1100 [1:44:56<44:54, 6.74s/it]
Loading weights [e6bb9ea85b] from /Users/ashutoshtyagi/Documents/Personal/gan/webui-sdxl/models/Stable-diffusion/sdXL_v10VAEFix.safetensors
Applying attention optimization: sub-quadratic... done.
Weights loaded in 25.5s (send model to cpu: 8.5s, load weights from disk: 0.7s, apply weights to model: 11.6s, move model to device: 4.6s).
100%|██████████████████████████| 25/25 [02:53<00:00, 6.94s/it]
100%|██████████████████████████| 25/25 [05:31<00:00, 13.25s/it]
100%|██████████████████████████| 25/25 [06:57<00:00, 16.71s/it]
100%|██████████████████████████| 25/25 [1:32:22<00:00, 221.69s/it]
Reusing loaded model sdXL_v10VAEFix.safetensors [e6bb9ea85b] to load talmendoxlSDXL_v11Beta.safetensors [7fb8947de2]█████████████████████████ | 800/1100 [3:34:04<53:37:16, 643.46s/it]
Loading weights [7fb8947de2] from /Users/ashutoshtyagi/Documents/Personal/gan/webui-sdxl/models/Stable-diffusion/talmendoxlSDXL_v11Beta.safetensors
Applying attention optimization: sub-quadratic... done.
Weights loaded in 45.0s (send model to cpu: 19.0s, load weights from disk: 1.5s, apply weights to model: 15.8s, move model to device: 8.6s).
100%|██████████████████████████| 25/25 [03:00<00:00, 7.21s/it]
100%|██████████████████████████| 25/25 [06:12<00:00, 14.91s/it]
100%|██████████████████████████| 25/25 [02:46<00:00, 6.68s/it]
100%|██████████████████████████| 25/25 [03:05<00:00, 7.41s/it]
Reusing loaded model talmendoxlSDXL_v11Beta.safetensors [7fb8947de2] to load wyvernmix15XL_xlV11.safetensors [1ba2448476]██████████████████████████████████████▎ | 900/1100 [3:51:42<22:30, 6.75s/it]
Loading weights [1ba2448476] from /Users/ashutoshtyagi/Documents/Personal/gan/webui-sdxl/models/Stable-diffusion/wyvernmix15XL_xlV11.safetensors
Applying attention optimization: sub-quadratic... done.
Weights loaded in 34.9s (send model to cpu: 18.5s, load weights from disk: 0.8s, apply weights to model: 11.2s, move model to device: 4.3s).
100%|██████████████████████████| 25/25 [02:59<00:00, 7.18s/it]
100%|██████████████████████████| 25/25 [06:07<00:00, 14.70s/it]
100%|██████████████████████████| 25/25 [03:04<00:00, 7.40s/it]
100%|██████████████████████████| 25/25 [02:58<00:00, 7.15s/it]
Reusing loaded model wyvernmix15XL_xlV11.safetensors [1ba2448476] to load xxmix9realisticsdxl_testV20.safetensors [ecb947b4ea]███████████████████████████████████████████████▋ | 1000/1100 [4:09:27<12:33, 7.54s/it]
Loading weights [ecb947b4ea] from /Users/ashutoshtyagi/Documents/Personal/gan/webui-sdxl/models/Stable-diffusion/xxmix9realisticsdxl_testV20.safetensors
Applying attention optimization: sub-quadratic... done.
Weights loaded in 27.9s (send model to cpu: 9.1s, load weights from disk: 0.8s, apply weights to model: 13.3s, hijack: 0.2s, move model to device: 4.4s).
100%|██████████████████████████| 25/25 [02:51<00:00, 6.86s/it]
100%|██████████████████████████| 25/25 [05:22<00:00, 12.89s/it]
100%|██████████████████████████| 25/25 [02:44<00:00, 6.59s/it]
100%|██████████████████████████| 25/25 [02:44<00:00, 6.56s/it]
Reusing loaded model xxmix9realisticsdxl_testV20.safetensors [ecb947b4ea] to load cherryPickerXL_v20.safetensors [e250118276]████████████████████████████████████████████████████████████████| 1100/1100 [4:24:40<00:00, 6.44s/it]
Loading weights [e250118276] from /Users/ashutoshtyagi/Documents/Personal/gan/webui-sdxl/models/Stable-diffusion/cherryPickerXL_v20.safetensors
Applying attention optimization: sub-quadratic... done.
Weights loaded in 24.2s (send model to cpu: 5.8s, load weights from disk: 0.9s, apply weights to model: 13.2s, move model to device: 4.3s).
Total progress: 100%|██████████████████████████| 1100/1100 [4:25:43<00:00, 14.49s/it]
Total progress: 100%|██████████████████████████| 1100/1100 [4:25:43<00:00, 6.44s/it]
And this is my system configuration:
Model Name: MacBook Pro
Model Identifier: Mac14,10
Chip: Apple M2 Pro
Total Number of Cores: 12 (8 performance and 4 efficiency)
Memory: 32 GB
System Firmware Version: 8422.121.1
OS Loader Version: 8422.121.1
OS Version: Ventura 13.4.1 (c) (22F770820d)
Please provide feedback.