Détails
Recommandé
Dark Kawaii - drkkaw-V2
V1
Dark Kawaii - drkkaw

Dark Kawaii - drkkaw

55
3
57
#Style

I would like to say thank you so much to MetroCat for helping me out with V2 and giving me some pointers on making it better I think my future Loras will come out way better because of it.

Dark Kawaii - drkkaw

I know its a long description if you don't like it don't read it. XD!

Prompt Used weights varied on tests.

Science Fiction, couch, <lora:drkkaw_epoch_10:0.8>,

I forgot to take out the negative prompt with the weights test but took the negative prompt out for the rest of the tests.

Negative prompt: JuggernautNegative

https://civitai.com/models/81563/juggernaut-negative-embedding?modelVersionId=86553

Sampler, Scheduler, Seed, and CFG are not important but I thought I would list them just incase you are curious.

Sampler: DPM++ 2M SDE

Scheduler: Karras

Seed: 586712160

CFG: 5

WEIGHTS!

Weights In this one came out tricky maybe you would say it's undertrained but I still think it came out good just need to raise weights instead of lowering I find 0.7 to 1.1 looks best but eye of the beholder and all that.

MODELS!

I find it is very picky with the models it works with which sucks because I love the PixelAhuskay model and I don't think it works well with it

Models tested "weights 0.7-1.1"

animeArtDiffusionXL_alpha2.safetensors [87f2e27ed8]

https://civitai.com/models/117259?modelVersionId=127055

realmixXL_v15.safetensors [ebf42d1fae]

https://civitai.com/models/124419?modelVersionId=136753

pixelAhuskay_v20.safetensors [1ef4aab75b]

https://civitai.com/models/537844/pixel-ahuskay

dreamshaperXL10_alpha2Xl10.safetensors [0f1b80cfe8]

https://civitai.com/models/112902?modelVersionId=126688

juggernautXL_version6Rundiffusion.safetensors [1fe6c7ec54]

https://civitai.com/models/133005?modelVersionId=198530

Trigger words?!?

The test above were used with the prompt

Science Fiction, couch, - then the lora and weight adjustments

this image shows with out the word Science Fiction.

Tags!

in automatic 1111 if you click on the lora tab.

Find the lora and click on the hammer and wrench.

Scroll down and you will find tabs.

Tags for your copy paste ease

robot, mecha, science fiction, glowing, no humans, realistic, non-humanoid, cable, standing, solo, one-eyed, radio antenna, spikes, weapon, antennae, lights, size difference, glowing eyes, gears, pink eyes, machinery, smoke, helmet, steampunk,

You can see how finding the tag words and testing them can change the image dramatically.

Prompt Problems?!?

Nice but lets say I was looking for more of a boy in the picture.

Amazing painting, boy, 10k ultra resolution, Science Fiction, weapon, <lora:drkkaw_epoch_10:0.8>

Steps: 35, Sampler: DPM++ 2M SDE, Schedule type: Karras, CFG scale: 5, Seed: 3187266996, Size: 1024x1224, Model hash: 53bb4fdc63, Model: animeArtDiffusionXL_alpha3, Lora hashes: "drkkaw_epoch_10: 98f8993d5274", Downcast alphas_cumprod: True, Version: v1.9.4

I put tag words I found in the instructions above in the negative prompt.

Amazing painting, boy, 10k ultra resolution, Science Fiction, weapon, <lora:drkkaw_epoch_10:0.8>

Negative prompt: robot, mecha, realistic, cable, one-eyed, radio antenna, spikes, antennae, lights, size difference, glowing eyes, gears, pink eyes, machinery, smoke, helmet, steampunk,

Steps: 35, Sampler: DPM++ 2M SDE, Schedule type: Karras, CFG scale: 5, Seed: 3187266996, Size: 1024x1224, Model hash: 53bb4fdc63, Model: animeArtDiffusionXL_alpha3, Lora hashes: "drkkaw_epoch_10: 98f8993d5274", Downcast alphas_cumprod: True, Version: v1.9.4

But lets say I wanted something a little more like a boy?!? leave the negative tags and lower the lora weight.

Amazing painting, boy, 10k ultra resolution, Science Fiction, weapon, <lora:drkkaw_epoch_10:0.6>

Negative prompt: robot, mecha, realistic, cable, one-eyed, radio antenna, spikes, antennae, lights, size difference, glowing eyes, gears, pink eyes, machinery, smoke, helmet, steampunk,

Steps: 35, Sampler: DPM++ 2M SDE, Schedule type: Karras, CFG scale: 5, Seed: 3187266996, Size: 1024x1224, Model hash: 53bb4fdc63, Model: animeArtDiffusionXL_alpha3, Lora hashes: "drkkaw_epoch_10: 98f8993d5274", Downcast alphas_cumprod: True, Version: v1.9.4

A friend gave me some pointers on ways that might make it better so I'm a try those and maybe have a newer version out by thee end of this weekend.

I talk a bit about it here I don't edit it so it's not safe to listen to at work or around family church wont like it that's for sure.

Voir la traduction

Notes & Commentaires

-- /5
0 Notes

Pas encore reçu suffisamment d'évaluations ou de commentaires

no-data
Aucune donnée disponible
avatar
Chatter avec le modèle
Annonce
2024-07-11
Publier un modèle
2024-07-14
Mettre à jour les informations du modèle
Détails du modèle
Type
LORA
Temps de Publication
2024-07-14
Modèle Basique
SDXL 1.0
Mots Déclencheurs
drkkaw
kawaii
science fiction
robot
mecha
Copier
Introduction de version

I would like to say thank you to MetroCat for giving me some pointers on making it a little better.

Weights are a little better in this one. Have fun and all that. I put drkkaw as a trigger word but didn't see it ad much when generating maybe you will ahve more luck though.

Périmètre de la licence
Source: civitai

1. Modèle partagé uniquement à l'apprentissage et au partage. Droits d'auteur et interprétation finale réservés à l'auteur original.

2. Auteur souhaitant revendiquer le modèle : Contactez officiellement SeaArt AI pour l'authentification. Nous protégeons les droits de chaque auteur. Cliquer pour revendiquer

Périmètre de la licence de création
Génération d'images en ligne
Effectuer une fusion
Autoriser le téléchargement
Périmètre de la licence de commerce
Les images générées peuvent être vendues ou utilisées à des fins commerciales
La revente ou la vente après fusion du modèle est autorisée.
QR Code
Télécharger l'App SeaArt
Poursuivez votre voyage de création AI sur mobile