Even_Adder

joined 1 year ago
 

(pelagic) (2024)

Image description: A cheerful young woman wearing a pink bunny onesie, sitting on a blue seat. The onesie features long ears, and she is making bunny ears with both hands near their face. A plush bunny toy sits beside her, mirroring her pose.

Full Generation Parameters:

score_9, score_8_up,score_7_up,1girl,solo,bunny pose,arms up,looking at viewer,(smile:0.6),(embarrassed:0.2),indoors,couch,open mouth, <lora:mochidaarisa_ponyXLV6:0.8>,cgmar,brown hair,long hair,low ponytail,hair scrunchie,brown eyes,pink bunny hood

Negative prompt: realistic,lowres,worst quality,bad anatomy,bad hands,high-contrast,monochrome

Steps: 30, CFG scale: 7, Sampler: Euler a, Seed: 3287128779, Size: 896x1152, Model: ponyDiffusionV6XL, Version: v1.7.0, Model hash: 67ab2fd8ec, Hires steps: 10, VAE Encoder: TAESD, Hires upscale: 2, Hires upscaler: 4x_fatal_Anime_500000_G, Variation seed: 1101, Denoising strength: 0.4, Variation seed strength: 0.11, Clip skip: 2

 

(OdiousStooge) (2024)

Image description: A large tortoise in the heart of a dense jungle. Green light filters through the canopy above onto the tortoise’s orange shell, providing a stark contrast against its dark blue-green skin.

Full Generation Parameters:

A vast, overgrown jungle where a giant turtle makes its way through the dense foliage. The turtle's massive shell is a miniature ecosystem, covered in giant, bioluminescent mushrooms that emit a soft, ethereal glow. Birds and small creatures nest among the fungi, adding life and movement to the scene. The turtle's slow, deliberate steps cause the ground to tremble slightly, and the glowing mushrooms light up its path through the dark, verdant jungle, in the style of decaying animatronic, claymation, VHS screen grab, grainy, old footage, in the style of an old 80's VHS dark fantasy, VHS screen grab, grainy, old footage , the scene is captured in dimly lit dark fantasy but vibrant colors, with bold ink lines defining form against the watercolor wash of the aged paper

Negative prompt: BadDream, worst quality, low quality, normal quality, monochrome, grayscale, (navel, lace), text, signature, watermark

Steps: 37, CFG scale: 3.5, Sampler: DPM++ 2M Karras, Seed: 352905922, Size: 832x1216, Created Date: 2024-06-23T2041:17.9999258Z, Clip skip: 2

 

(dude_) (2024)

Image description: A young woman with short dark hair, sitting in a tranquil forest setting. She is dressed in a sleeveless dress and is positioned with her hands resting on the knees. Dappled sunlight filters through the canopy of green leaves above, casting patterned shadows across the scene and the character. The character gazes off to the side with a neutral expression, suggesting a moment of introspection or quiet contemplation.

Full Generation Parameters:

<lora:takano_akira_ponyxl_v2:0.9>, takira, short hair, dark red hair, brown hair, purple eyes, half-closed eyes, solo, looking at viewer, (giraffe costume:1.2), animal costume, outdoors, sitting on tree stump, dappled sunlight, tree, tree shade, large tree, foliage, grass, bush, nature, backlighting, score_9, score_8_up, score_7_up, score_6_up, anime coloring, uncensored, <lora:UrushiharaSatoshi_XL_PONY_V2:0.4>

Negative prompt: line art, watermark, logo, (worst quality:1.5), (low quality:1.5), (normal quality:1.5), lowres, bad anatomy, bad hands, multiple eyebrow, (cropped), extra limb, missing limbs, deformed hands, long neck, long body, signature, username, artist name, conjoined fingers, deformed fingers, error, (deformed|distorted|disfigured), poorly drawn, wrong anatomy, mutation, mutated, (mutated hands AND fingers), bad fingers, loss of a limb, extra limb, missing limb, floating limbs, amputation, deformed, black and white, disfigured, low contrast

Steps: 28, CFG scale: 7, Sampler: Euler a, Seed: 2017541030, VAE: ponyDiffusionV6XL_v6StartWithThisOne.vae.safetensors, Size: 1040x1520, Model: ponyDiffusionV6XL_v6StartWithThisOne, Model hash: 67ab2fd8ec, kohya_hrfix_enabled: True, kohya_hrfix_end_percent: 0.35, kohya_hrfix_block_number: 3, kohya_hrfix_start_percent: 0, kohya_hrfix_upscale_method: bicubic, kohya_hrfix_downscale_factor: 1.5, kohya_hrfix_downscale_method: bicubic, kohya_hrfix_downscale_after_skip: True

28
Gentle Cast (image.civitai.com)
 

(jice) (2024)

Image description: A man, dressed in neutral tones and wearing a hat, stands at the edge of a calm stream dotted with blooming water lilies, engrossed in fishing. The surrounding landscape is lush with greenery, and the sunlight filtering through the trees casts a soft glow on the scene.

Full Generation Parameters:

a fisherman,lake,cristal clear water,beautiful trees,beautiful flowers,blue light,romanticism art,(landscape art stylized by Karol Bak:1.3),Paul Gauguin,Cyberpop,short lighting,F/1.8,extremely beautiful,<lora:oil-painting:2>, oil painting of

Steps: 10, CFG scale: 2, Sampler: Euler, Seed: 1784836563, Size: 900x1200, Model: CreaPrompt_Ultimate3, Version: v1.9.4, Model hash: c83f76acb6, Schedule type: Uniform,

ADetailer model: face_yolov8n.pt, ADetailer version: 24.6.0, ADetailer x offset: 4, ADetailer y offset: -2, ADetailer mask blur: 8, ADetailer model 2nd: mediapipe_face_mesh_eyes_only, ADetailer confidence: 0.44, ADetailer dilate erode: 4, ADetailer mask blur 2nd: 4, ADetailer confidence 2nd: 0.3, ADetailer inpaint padding: 32, ADetailer dilate erode 2nd: 4, ADetailer denoising strength: 0.36, ADetailer inpaint only masked: True, ADetailer inpaint padding 2nd: 32, ADetailer denoising strength 2nd: 0.49, ADetailer inpaint only masked 2nd: True, ADetailer mask only top k largest: 1

[–] [email protected] 3 points 18 hours ago

How to safely move a live wire.

4
Nanako - Pokémon (image.civitai.com)
 

(TecnoIA) (2024)

Image description: A girl with blue hair striking an enthusiastic pose is dressed in an outfit consisting of a white cap with a yellow symbol, a yellow-striped jacket, a blue t-shirt, and striped shorts.

Full Generation Parameters:

score_9, score_8_up, score_7_up, score_6_up, score_5_up, score_4_up, BREAK source_anime, anime screencap, <lora:CaseyPkmnXL-08:0.7> CaseyPkmnXL, 1girl, solo, striped shorts, clenched hand, open mouth, hat, shirt, blue hair, eyelashes, baseball cap, striped, smile, shorts, pink background, jacket, :d, knees, tongue, vertical stripes, looking at viewer, outstretched arm, open clothes, open jacket

Negative prompt: score_6, score_5, score_4, holding, monochrome, simple background, 3d,

Steps: 8, CFG scale: 3, Sampler: DPM++ 2S a, Seed: 703577234, Size: 896x1152, Model: autismmixSDXL_autismmixLightning, Version: f0.0.17v1.8.0rc-latest-276-g29be1da7, Model hash: f1f226aa36, Clip skip: 2

 

(LordTerror) (2024)

Image description: A lone figure standing on the edge of a cliff overlooking a misty seascape. Above them towers an imposing tree with vibrant red foliage that contrasts starkly against the surrounding subdued tones.

Full Generation Parameters:

by (((Nathan Wirth) and Thomas Cole) and Becky Cloonan) and James Gilleard, surreal art, full body length, cinematic , hyperdetailed photography, , shallow depth of field, <lora:more_art:0.40><lora:faetastic:0.25><lora:midjourney:0.15>

Negative prompt: low quality, bad anatomy, mess, bizarre, cartoon, painting, illustration

Steps: 24, CFG scale: 7, Sampler: DPM++ 2M Karras, Seed: 2142316041, VAE: sdxl_vae.safetensors, Size: 768x1152, Model: zavychromaxl_v40, Version: v1.7.0, Template: [object Object], Model hash: 63a3752da1, Hires steps: 10, Hires upscale: 1.4, Hires upscaler: 4x-UltraSharp,

ADetailer model: face_yolov8n.pt, ADetailer prompt: [object Object], ADetailer VAE 3rd: vae-ft-mse-840000-ema-pruned.ckpt, ADetailer version: 24.1.2, Negative Template: [object Object], Denoising strength: 0.5, ADetailer mask blur: 12, ADetailer model 2nd: hand_yolov8n.pt, ADetailer model 3rd: female_breast_v3.2.pt, ADetailer steps 3rd: 20, ADetailer confidence: 0.5, ADetailer prompt 2nd: __prompt__ __detailer/hands__, ADetailer prompt 3rd: __prompt__ __detailer/breasts__, ADetailer sampler 3rd: DPM++ 2M Karras, Hires negative prompt: [object Object], ADetailer dilate erode: 12, ADetailer CFG scale 3rd: 7.0, ADetailer inpaint width: 1024, ADetailer mask blur 2nd: 12, ADetailer mask blur 3rd: 12, ADetailer checkpoint 3rd: SD1.5\wafflemix7.safetensors [4844f773bb], ADetailer confidence 2nd: 0.5, ADetailer confidence 3rd: 0.5, ADetailer inpaint height: 1024, ADetailer mask min ratio: 0.001, ADetailer inpaint padding: 256, ADetailer negative prompt: [object Object], ADetailer dilate erode 2nd: 12, ADetailer dilate erode 3rd: 12, ADetailer inpaint width 2nd: 896, ADetailer inpaint width 3rd: 1024, ADetailer denoising strength: 0.45, ADetailer inpaint height 2nd: 896, ADetailer inpaint height 3rd: 1024, ADetailer mask min ratio 2nd: 0.001, ADetailer mask min ratio 3rd: 0.001, ADetailer inpaint only masked: True, ADetailer inpaint padding 2nd: 256, ADetailer inpaint padding 3rd: 256, ADetailer negative prompt 2nd: [object Object], ADetailer negative prompt 3rd: [object Object], ADetailer use separate VAE 3rd: True, ADetailer denoising strength 2nd: 0.35, ADetailer denoising strength 3rd: 0.45, ADetailer use separate steps 3rd: True, ADetailer inpaint only masked 2nd: True, ADetailer inpaint only masked 3rd: True, ADetailer mask only top k largest: 5, ADetailer use inpaint width height: True, ADetailer use separate sampler 3rd: True, ADetailer use separate CFG scale 3rd: True, ADetailer mask only top k largest 2nd: 6, ADetailer mask only top k largest 3rd: 4, ADetailer use separate checkpoint 3rd: True, ADetailer use inpaint width height 2nd: True, ADetailer use inpaint width height 3rd: True

 

Highlights for 2024-06-23

Following zero-day SD3 release, a 10 days later here's a refresh with 10+ improvements
including full prompt attention, support for compressed weights, additional text-encoder quantization modes.

But there's more than SD3:

  • support for quantized T5 text encoder FP16/FP8/FP4/INT8 in all models that use T5: SD3, PixArt-Σ, etc.
  • support for PixArt-Sigma in small/medium/large variants
  • support for HunyuanDiT 1.1
  • additional NNCF weights compression support: SD3, PixArt, ControlNet, Lora
  • integration of MS Florence VLM/VQA Base and Large models
  • (finally) new release of Torch-DirectML
  • additional efficiencies for users with low VRAM GPUs
  • over 20 overall fixes
 

(Manityro) (2024)

Image description: A woman with cat-like features standing in a moonlit forest. She is holding an ornate staff, topped with a glowing purple orb and adorned with smaller gems of the same color. Dressed in a dark, feathered outfit, she exudes an air of mystique. The full moon casts a soft glow on the forest and the character.

Full Generation Parameters:

score_9, score_8_up, score_7_up, 1girl, solo, shtla, dark-skinned female, animal ears, tail, whisker markings, short hair, large breasts, <lora:YshtolaRhulPDXL_V1-Manityro-CAME:1.0>, grey eyes, white hair, shaOtft, feather earrings, feathers, brooch, black halterneck, black choker, black dress, long dress, fur-trimmed dress, wide sleeves, black gloves, partially fingerless gloves, claw ring, looking at viewer, serious, holding staff, wood staff, outdoors, dark, night, moon, spooky forest, purple magic, purple aura

Negative prompt: monochrome, text, signature, artist name, watermark, patreon, moles, mole, pubes, tanlines,

Steps: 24, CFG scale: 6, Sampler: Euler A AYS, Seed: 3622385744, VAE: sdxlVAE_sdxlVAE.safetensors, Size: 832x1216, Model: snowpony_v10, Version: f0.0.17v1.8.0rc-latest-287-g77bdb920, Model hash: d6f941b46b, Hires steps: 10, Hires upscale: 1.5, Hires upscaler: 4x-UltraSharp,

ADetailer model: Anzhc Face seg 640 v2 y8n.pt, ADetailer version: 24.6.0, Denoising strength: 0.4, ADetailer mask blur: 8, ADetailer confidence: 0.3, ADetailer dilate erode: 4, ADetailer inpaint width: 1024, ADetailer inpaint height: 1024, ADetailer inpaint padding: 32, ADetailer denoising strength: 0.4, ADetailer inpaint only masked: True, ADetailer use inpaint width height: True, Clip skip: 2

19
Twilight Gazer (image.civitai.com)
 

(CivitFowl) (2024)

Image description: A fantastic stag standing in a moonlit forest. Its body reflects the glow of the bioluminescent plants around it, and a puddle mirrors its striking form below.

Full Generation Parameters:

(8k digital photography, raw photo:1.2), a enchanted beautiful summer landscape at (dawn with a fantastical creature, magical lights in the sky:1.1) <lora:Bio-Luminescence:1> bioluminescent

Steps: 7, CFG scale: 3, Sampler: DPM++ SDE Karras, Seed: 2368414684, VAE: sdxl_vae.safetensors, Size: 612x1024, Model: stableFusionXLTurbo_v161, Version: f0.0.17v1.8.0rc-latest-276-g29be1da7, Model hash: 913421be31, Hires steps: 7, Hires upscale: 1.5, Hires upscaler: 4X-NMKD.pth, Denoising strength: 0.7

[–] [email protected] 3 points 1 day ago

I'm just wondering if it'll be Open Source. It seems like they're working for Nvidia now.

 
 

(HellKaiza) (2024)

Image description: A woman with wild pink hair wearing a black top and white puffy pants. She is holding out their hand, pointing her finger, with a glowing pink orb of energy floating in front of it.

Full Generation Parameters:

source_anime, score_9, score_8_up, score_7_up, anime screencap, 1girl, death beam, solo, long hair, [[[gold choker]]], breasts, looking at viewer, smile, navel, cleavage, earclip, medium breasts, majin tail, white hair, detached sleeves, pointy ears, midriff, collarbone, majin eyes, hoop earrings, black tube top, pink skin, energy ball, majin pants, android 21, arm ring, majin, <lora:android21_pony_v1:0.8>

Negative prompt: score_6, score_5, score_4, censored, close up,, low quality , worst quality, bad anatomy, bad proportions, extra legs, deformed anatomy, messy color, deformed fingers, bad, distracted, extra arms, blurry, extra finger, stone, mountains, cover, bad drawn fingers, extra digits, extra toes,

Steps: 45, CFG scale: 7, Sampler: Euler a, Seed: 3499731711, VAE: sdxl_vae.safetensors, ENSD: 31337, Size: 640x960, Model: autismmixSDXL_autismmixPony, Version: v1.6.0-2-g4afaaf8a, Pad conds: True, Model hash: 821aa5537f, Mimic scale: 22.5, Hires upscale: 2, Hires upscaler: 4x-AnimeSharp,

`ADetailer model: face_yolov8n.pt, Interpolate Phi: 1, ADetailer version: 24.4.2, Denoising strength: 0.1, Scaling Startpoint: MEAN, ADetailer mask blur: 4, Variability Measure: AD, ADetailer confidence: 0.3, Threshold percentile: 100, ADetailer dilate erode: 4, ADetailer inpaint padding: 32, Separate Feature Channels: True, ADetailer denoising strength: 0.4, Dynamic thresholding enabled: True, ADetailer inpaint only masked: True

 

(LordTerror) (2024)

Image description: A solitary figure stands atop a rocky outcrop, reaching towards a brilliant beam of light that pierces the night sky, emanating from what appears to be a galaxy above. The sky transitions from twilight at the horizon to the dark expanse of space, sprinkled with stars.

Full Generation Parameters:

by [Cyril Rolando:Peter Sculthorpe:0.25] and [[Gustave Baumann|Tomasz Alen Kopera]:Gustave Baumann and Tomasz Alen Kopera:0.65] [, landscape::1]

Negative prompt: Unspeakable-Horrors-Composition-SDXL, SDXL_TI_my_eyes_are_bleeding_2.0, unaestheticXL_Alb2, unprofessional, messy, sketch, deformed, painting, splotchy, topless

Steps: 30, CFG scale: 4, Sampler: DPM++ 3M SDE Karras, Seed: 3023833774, VAE: sdxl_vae.safetensors, Size: 768x1152, Model: newrealityxlAllInOne_Newreality40, Version: v1.8.0, Model hash: 32b09fe763, Hires steps: 12, Hires upscale: 1.5, Hires upscaler: 4x-UltraSharp,

ADetailer model: face_yolov8n.pt, ADetailer prompt: [object Object], ADetailer VAE 3rd: vae-ft-mse-840000-ema-pruned.ckpt, ADetailer version: 24.3.0, Denoising strength: 0.45, ADetailer mask blur: 12, ADetailer model 2nd: hand_yolov8n.pt, ADetailer model 3rd: female-breast-v4.0-fantasy.pt, ADetailer steps 3rd: 20, ADetailer confidence: 0.5, ADetailer prompt 2nd: [object Object], ADetailer prompt 3rd: [object Object], ADetailer sampler 3rd: DPM++ 2M Karras, ADetailer dilate erode: 12, ADetailer CFG scale 3rd: 7.0, ADetailer inpaint width: 1152, ADetailer mask blur 2nd: 12, ADetailer mask blur 3rd: 12, ADetailer checkpoint 3rd: SD1.5\wafflemix7.safetensors [4844f773bb], ADetailer confidence 2nd: 0.5, ADetailer confidence 3rd: 0.4, ADetailer inpaint height: 1152, ADetailer mask min ratio: 0.001, ADetailer inpaint padding: 256, ADetailer dilate erode 2nd: 12, ADetailer dilate erode 3rd: 12, ADetailer inpaint width 2nd: 896, ADetailer inpaint width 3rd: 1024, ADetailer denoising strength: 0.45, ADetailer inpaint height 2nd: 896, ADetailer inpaint height 3rd: 1024, ADetailer mask min ratio 2nd: 0.001, ADetailer mask min ratio 3rd: 0.001, ADetailer inpaint only masked: True, ADetailer inpaint padding 2nd: 256, ADetailer inpaint padding 3rd: 256, ADetailer negative prompt 3rd: [object Object], ADetailer use separate VAE 3rd: True, ADetailer denoising strength 2nd: 0.45, ADetailer denoising strength 3rd: 0.45, ADetailer use separate steps 3rd: True, ADetailer inpaint only masked 2nd: True, ADetailer inpaint only masked 3rd: True, ADetailer mask only top k largest: 5, ADetailer use inpaint width height: True, ADetailer use separate sampler 3rd: True, ADetailer use separate CFG scale 3rd: True, ADetailer mask only top k largest 2nd: 6, ADetailer mask only top k largest 3rd: 4, ADetailer use separate checkpoint 3rd: True, ADetailer use inpaint width height 2nd: True, ADetailer use inpaint width height 3rd: True

[–] [email protected] 4 points 4 days ago

Tiamat-8b-1.2-Llama-3-DPO-GGUF

[–] [email protected] 3 points 6 days ago* (last edited 6 days ago)

Have you tried any of the pixel fine-tunes on Civitai?

This one looks pretty good.

[–] [email protected] 2 points 6 days ago (4 children)

What models do you use?

[–] [email protected] 6 points 1 week ago

Image goes hard.

[–] [email protected] 9 points 1 week ago* (last edited 1 week ago)

If it's the same one from a few months ago, the wording is so vague that only huge companies with legal departments will be able to navigate the compliance maze they've set up.

https://www.eff.org/deeplinks/2023/07/generative-ai-policy-must-be-precise-careful-and-practical-how-cut-through-hype

[–] [email protected] 3 points 1 week ago (1 children)

At least it's not Bandai, but it's likely gacha since it's releasing on the AppStore, GooglePlay, and Steam.

[–] [email protected] 5 points 1 week ago

I saw an interesting video about image classifiers yesterday.

[–] [email protected] 1 points 1 week ago* (last edited 1 week ago)

Check out Civitai for Stable Diffusion models. I'm not quite sure which model they are using, but you may be able to find something on there.

Also, there's a Stable Diffusion community at [email protected]. If you make a thread there we can help you find what model they use or something similar.

[–] [email protected] 3 points 1 week ago (1 children)

SDNext can run Pixart-Σ.

[–] [email protected] 39 points 1 week ago (4 children)

The model should be capable of much better than this, but they spent a long time censoring the model before release and this is what we got. It straight up forgot most human anatomy.

view more: next ›