Please consider buying my two albums Sisters of the Wicked and Danger Ahead, my book Dracula and also please check out my YouTube channel. Civitai won't support creators, so any purchase is a real help. Thanks.
DRACULA https://www.amazon.com/dp/B0FM4DXYPL
Sisters of the Wicked https://amazon.com/music/player/albums/B0DSJY3ZHY?marketplaceId=ATVPDKIKX0DER&musicTerritory=US&ref=dm_sh_CdyeuxIBNBDOjZ1hpB1PPJxT0
Danger Ahead https://www.amazon.com/music/player/albums/B0DB2DV354
Music In The Machine https://www.youtube.com/@musicinthemachine/videos
AcornXL V4
https://www.youtube.com/@musicinthemachine/videos
I use my model in making of my YouTube music videos. Feel free to check it out.
V4 I have improved the skin realism. Improvements in hands. In my own generation I get good hands the majority of the time. Of course, it's still AI, so still get a deformed one sometimes.
I have changed by basic resolution to 1024 X 1328. It seems to be part of the hand improvement. My favorite sampler is still DPM++ 2M SDE Heun Exponential but many get great results. My preference is for 60 steps, but you can get good results with 20 or less as well. Usually CFG from 2.5 to 3.5 unless you are doing something unusual, then crank it up if you like.
AcornXL V3
Any follows help as well as checking out my YouTube channel or most of all any direct donations on my BuyMeACoffee.
https://buymeacoffee.com/wadebabcoc1
https://www.youtube.com/@musicinthemachine/videos
Similar to my Acorn Is Boning models, I made a Randomizer prompt for use with Dynamic Prompts. Nice if you want to generate a bunch of stuff without having to keep redoing prompts. Here is mine if you want to use it. Of course you can also modify it with different colors, clothes, places and whatever options you want.
a (high definition) photo of a 19 year old {(ginger:1.2) | (blonde:1.3) | (platinum blonde:1.5) | (brunette:1.3) | (redhead:1.3)} woman {(wearing {black | pink | red | blue} bikini:1.5) | (wearing {black | pink | red | blue} {crop | linen | spandex | lace} shirt and {jeans | skirt | {black | pink | red | blue} shorts | {black | pink | red | blue} yoga pants}:1.5) | (wearing sexy {black | yellow | purple | blue | red} dress, jewelry:1.5) | (naked:1.3){(hairy pussy:1.4) | (shaved pussy:1.4)}}, {(standing:1.4) | (with another woman, embracing:1.5) | (sitting:1.5)}, {petite | skinny | chubby}, {(large sagging breasts:1.4) | (medium breasts:1.3) | (small breasts:1.3)}, (ultra fine body hair), {on black sand beach | in college dorm room | in office for job interview | in a large van | in bedroom of fancy hotel | by the swimming pool | (in a crowded train station) | on a tropical beach | in a department store | on a skyscraper roof top | in a private jet | on the kitchen table | in the shower}, (perfect hands:1.5), (cinematic), (dark, high contrast), narrow hips, slim petite body, (intricately detailed eyes:1.5), real (skin textures), (subsurface scattering:0.9)
(cropped head:1.5), (extra arm:1.5), (extra fingers:1.5), (deformed hands:1.5), (missing fingers:1.5), (deformed feet:1.5), (deformed leg:1.5), (missing leg:1.5), (anime:1.4), (freckles:1.4), (watermark:1.5)
1024 X 1280
Steps: 60
CFG 2.5
DPM++ 2M SDE Heun Exponential
These are my typical setting, but the model is excellent with really any steps, sampler and CFGs that I have tried.
AcornXL V2
This is my first update for AcornXL. It's been a while, but I don't upload unless I believe I have made considerable improvement. I believe that is now the case.
https://www.youtube.com/@musicinthemachine/videos
I use some of my models for videos on my YouTube page if you want to check it out.
Acorn has always been primarily a photo model. It still is, which I think my samples will show, but I have included a variety of styles in my initial examples to show the model is not only among the best, if not THE best, photo model available, but also a versatile model capable of many styles.
There have been a lot of changes in Stable Diffusion since my first XL model. Updates to Automatic1111 which is what I primarily use, as well as others such as ComfyUI which I use some as well. I still get the best results in A1111. I find for photo quality images, it is best to use CFG in the 3.5 to 4.5 range. My favorite samplers are:
DPM++ 3M SDE Karras
DPM++ 2M Karras
and DPM++ 2M SDE Heun Exponential
Although good result can be achieved at around 20 steps, I find that taking a little extra time and doing 60 steps tend to have good payoffs.
AcornXL V1
This is my first XL model. It's definitely not perfect as there just isn't a lot to work with yet for SDXL. It is pretty good though. I use a mixture of samplers, but so far my favorites are DPM++ 2M Karras and DPM++ SDE Karras. Normally CFGs from 4.5 to 7. Steps between 20 and 40. I have experimented using it together with the base SDXL refiner and so far it seems to improve results. If you use the up to date version of Automatic1111 it allows you to have the refiner kick it at whatever percentage you want. Some of my example pic use it starting at 50% up to 90%. Just use some trial and error and see what works for you.
AND as always, my sample images use NO loras, textual inversions or negative TIs. Model and webui extensions only.
sdxl_vae is baked in
Feel free to post your gens down below. As always, I love to see what you make.
Acorn Photo Diversity V5
After releasing my Diversity V4 of the Acorn Photo models, I wasn't quite happy with the photo quality level. I kept testing and I believe this new version does a better job.
Prompting is different with this version. No longer is the (photorealistic) prompt needed and overall you often need less prompting and less negative prompts. As always, my sample images contain no Loras or TIs, either positive or negative.
Samplers I use mostly DPM++ 2M Karras and DPM++ 3M SDE Karras although many work fine, these are my favorites.
CFG anything from 6 down to 2 work good. I have found that if facial expression are too strong at 5 or 6, running it again at 2.5 or 3 still generates a great image but tones down exaggerated stuff.
Steps are usually 20 to 30.
vae-ft-mse-840000-ema-pruned baked in to all 1.5 models of the Acorn Series
Acorn Photo Diversity V4
Updated to generate a greater range of ethnicities.
Acorn Photo V3
It's getting hard to tell these generations are not actual photographs. Not sure how much more we will get out of SD1.5. For prompting, version 3 stays pretty much the same. I use DPM++ 2M Karras mostly, but also DPM++ SDE Karras and DPM++ 2M SDE Karras. CFGs from 4 to 6 mostly but depending on preference it doesn't get bad well up to 10.
Post your creations down below. I would love to see what you make.
Acorn Photo V2
This model increases the photo realism significantly over the previous version while still being a very versatile model. It is great for portrait type people pics but also for more action oriented things like Fantasy RPG type shots.
Post your creations down below. I would love to see what you make.
Similar to previous version, seems to help a lot to use (photorealistic:1.4) in the prompt.
I find it works best at CFGs from 4.5 to 7, but gets some good results at other levels.
As with a lot of models, the lower CLIP Skips seems best, but I have gotten very good results with 1,2 or 3 and good images that drift farther away from the prompt as you go higher.
As before I can't accurately state exactly what is in the mix or how much weight each has. A lot of trial and error testing. The list of models included that I am sure about are:
Photon_v10
Aamon_v10
Colorful_v31
ConsistentFactor_euclidV51
AlmostHumanMix_v10r
EroticVision_v4
NextPhoto_v20
LOFI
QGOPromptingreal_qgo10a
VirileReality
RPG
Acorn Photo
This is not a replacement of Acorn Is Spinning, but an addition with a little different use case. Feel free to play with both and see what you can make. Please post any examples you like as well.
My goal is to get a model that is versatile enough to generate any kind of content equally well. To date, this is the most photorealistic model I have made. Acorn Is Spinning seems slighty more versitile in what it will do, but Acorn Photo seems to be a little better at just pure photos of people.
I recommend using the term (photorealistic:1.4) in the prompt. Also 1.5 or 1.6 work well.
Recommended negative: anime, cartoon, paintings, glossy skin, (worst quality:2), (low quality:2), ((monochrome)), ((grayscale)), (skin tags:1.5), skin spots, acne, skin blemishes, Watermark, Text, deformed, disfigured, poorly drawn face, mutated, extra limb, poorly drawn hands, missing limb, floating limbs, disconnected limbs, disconnected head, malformed hands, long neck, mutated hands and fingers, bad hands, missing fingers, cropped, huge calf, fused hand
The examples I upload to demonstrate the model are done with no Loras, Textual Inversions or LyCos. Nothing but the prompts and upscales.
Acorn Is Spinning
This model is intended to produce the best photo real images that I can get. I am no expert in models and this is a merge of multiple other models created by those who know a lot more than me. What I can say is that in using a lot of models, this merge has given me the best results I have achieved.
My true goal is to get a model that is versatile enough to generate any kind of content equally well. Most models are designed to make attractive women, which I certainly appreciate, but I would like to be able to just as easily generate several people, both men and women without having to fight it.
The examples I upload to demonstrate the model are done with no Loras, TIs or LyCos. Nothing but the prompts and upscales.
This model is obviously not perfect. I can't accurately state exactly what is in the mix or how much weight each has. The list of models included that I am sure about are:
Photon_v10
Aamon_v10
Colorful_v31
ConsistentFactor_euclidV51
AlmostHumanMix_v10r
EroticVision_v4
NextPhoto_v20
LOFI
QGOPromptingreal_qgo10a
There might be a few other in the mix, but the heaviest weight started with Photon. On it's own, it seems to be about the best single model I have found but mixing in some of the others made some big improvements to the overall outputs and versatility.
I can take no credit for creation, so if you like the model, enjoy it and mix in other new stuff and make it better.
I get the best results with CFGs from 5 to 6, with DPM++ 2S a Karras and DPM++ 2M Karras. Normally around 25 to 27 steps. Best to use the highest resolution you can as well. I do a lot at 1024 X 768 as well as 768 X 768 and recently started using the the ADetailer instead of the Restore Faces.
For Clip Skips, I have gotten some great results with all of them up to 12, but the best are probably 1,2,3 and 6 most of the time. It seems to get nice results all the way up, but the higher you go, the farther it gets from the prompt while still giving pleasing images. Just maybe not the one you were looking for.
Description
Not sure how much better models will be able to get for 1.5. Most of the generations at least of people are indistinguishable from photos or at very nearly so. Obviously there could still be improvement in things like hands and feet.
FAQ
Comments (17)
If 1.5 is to continue developing, it's not going to be about more realism or detail, it's going to be about training in new data to get more variation and unique styles. So many models are exactly the same these days, mixes are getting to the point where they just make the same renders even between models.
I like a few of the newer models done by photographers and photo enthusiasts who train their own work in, especially on XL. This will add character and uniqueness to the renders, and make for more artful generations. PAWG nudes and waifus are definitely done at this point, maybe we can actually find creativity now.
They're kind of cannibalizing themselves. Some of my favorite realism models from the beginning are progressively getting worse.
@SyntheticSunsets I'm not sure which models you are talking about with realism models. For me realism requires things actually looking real. As far as I am concerned this model and the newest epiCDream and a couple others are the only ones that just now are reaching photo quality. Most of the model that call themselves realism look like anime or pixar.
@rbbcck381 epicMystery, and now epicDream are my current favorites. Maybe that is why the others kind of pale in comparison to me now.
You're really shooting yourself in the foot by offering this as an early access model. The samples aren't notable, at least not in the way that this model is unique or otherwise inclusive. You're sacrificing your exposure at the top of the 'newest model' feed.
That last for about 20 minutes anyway. It's not like I get paid for it. I am not concerned with being unique. The only reason I do it is to work toward photo realism. My generations look like photos and that is what matters to me.
@Seeker70 understood. I don't mean to be critical of your model, just stating that you're robbing yourself of exposure at the expense of the obvious implications of the 'early access' feature.
@SyntheticSunsets agreed on the "Early Access" situation. I always browse by newest models, and when I see a model behind an EA tag I always forget to check on it by the time the time expires and I guess I miss out on downloading it as I forget it even exists.
Other than that, this looks like a good quality model though, so kudos to @Seeker70 on the model itself! Hopefully I don't forget to come back to this one when the timer's up
@Gairm I don't usually do the EA thing. Every time I go to upload a model it asks about doing and I just said why not. Yeah, I probably should have skipped it. I typically browse by newest models myself as well. I don't care too much about exposure, as I don't get anything for it, but normally I just upload and post right away. I just did a new version of my Acorn Is Boning model too which I just posted too. I suppose if there is a next model, I will likely just post it right away too.
@SyntheticSunsets You're probably right. I should have just posted it like I usually do. The model itself is awesome, at least for my needs. For the most part I only care about photorealism. I don't do anime or any of that stuff. It took quite a while to do this model because V2 was so good that anything I tried merging with it degraded it rather than improving it. I think this one is improved on just outright photorealism, but a bit of a trade off is it is not quite as good at varying faces as V2. On V2 I could pretty regularly just say 3 women, different faces and they would be distinctly different. This one on just raw photo look is a little better, but not quite as easy to get completely different faces like that.
@Seeker70 We appreciate your work, and this one does seem good so far. I will have to do some more with it and make a review with some pics. By the way, anyone can select the "Notify Me" link on these EA releases to get a notification when it goes public. I was annoyed at these for a while, but at this point I just hit the link and wait for the notification. Not like I don't have enough to play with at this point!
I apologize if that came off as blunt or condescending. I am in digital marketing and sincerely just trying to offer advice. I leave the same comment on any worthwhile model (it obviously caught my attention) because I also check in regularly and browse 'newest models' first. By 'not notable', I am speaking from the perspective of someone that's been around a while.. so for me notable is something breakthrough, which aside from a few, has been far and few the past two months or so with 1.5
@SyntheticSunsets No problem. I have been watching AI for a while myself and generally I am into all kinds of technology. When it comes to models, the only thing I care about at all is how close it can come to be photo real. Even though I am into fantasy art, my goal is fantasy art that looks as though it is a photograph of something that doesn't currently exist. I have yet to see any model at all that makes anything that looks real other than people. As far as I am concerned there really hasn't been any breakthrough models yet. I have yet to see a single model that can reliably display a warrior holding a sword that actually looks like a sword. I hope we eventually get there.
great model, mate! Getting by far best results compared with similar models. Really appreciate your work!
Thanks. I use it myself as well. I won't say it is necessarily the best model, but in doing experiments to continue to improve it, I have a lot of trouble finding any new potential merge that doesn't degrade it. Part of why I don't spend a lot of time at the top of the new models page. I don't do updates just to be at the top of the page. Glad you are enjoying it.
One the best photorealistic merges out there! Currently my #1 for sure. Very responsive and promptable, great job!
Yes, it is what I use most of the time too. I always like to try others, but so far, I have always stuck with this one as I have yet to find anything that is better. I am glad you are enjoying it.