<h1 id="deepcobalt">[DeepCobalt]</h1>Recommended Parameters:<ul><li>Sampler: Euler a</li><li>CFG Scale: 3~4</li><li>Steps: 20~28</li><li>Clip Skip : 2</li></ul><h1 id="deepnavy">[DeepNavy]</h1>Recommended Parameters:<ul><li>Sampler: DPM++_2M_SDE</li><li>CFG Scale: 4</li><li>Steps: 24</li><li>Clip Skip : 2</li></ul><h1 id="deepblue">[DeepBlue]</h1>Recommended Parameters:<ul><li>Sampler: Euler a</li><li>CFG Scale: 3 – 5</li><li>Steps: 20 – 30</li><li>Clip Skip : 2</li></ul><h1 id="deepviolet">[DeepViolet]</h1>Vpred Model

Mol_Keun Mix

I worked on removing the noise that used to appear around the eyes and nose in previous versions. As a result, there are quite a few differences from the original DeepBlue model, but I feel that this version understands prompts much better. You can now use the Euler a sampler without getting any of that noise.

molKeunMix_deepnavyV1EPS.safetensors

DeepNavy V1 EPS

DeepCobalt V2

DeepCobalt

DeepNavy V4

DeepNavy V2 EPS

DeepBlue V2 EPS

DeepViolet VPR

DeepBlue EPS

V3 [NoobAI-XL]

V2 [NoobAI-XL]

V1 [NoobAI-XL]

V0.1 [ILXL 0.1]

V2 [Pony]

V1 [Pony]

V1 [SD 1.5]

V0.1 [SD 1.5]

V6.0

v4.5

v5.0

v4.0

v3.0

v2.0

v1.0

Previous name: Mol_Keun Furry Mix / CuteCrunchcreator: Mol_Keun<a target="_blank" rel="ugc" href="https://civitai.com/models/2028510/deepnavy">[DeepNavy V3 Updated Here]</a>[DeepNavy]I worked on removing the noise that used to appear around the eyes and nose in previous versions. As a result, there are quite a few differences from the original DeepBlue model, but I feel that this version understands prompts much better. You can now use the Euler a sampler without getting any of that noise.Recommended Parameters:<ul><li>Sampler: DPM++_2M_SDE</li><li>CFG Scale: 4</li><li>Steps: 24</li><li>Clip Skip : 2</li></ul>Previous Parameters:<ul><li>Sampler: Euler a</li><li>CFG Scale: 3 – 5</li><li>Steps: 20 – 30</li><li>Clip Skip : 2</li></ul>[DeepBlue V2]This is a merged model using a different base model instead of the original(V1~V3) model. I tried to keep the feel as similar as possible, but since the model and LoRAs are entirely different, I adjusted the weights one by one during the merge.Recommended Parameters:<ul><li>Sampler: DPM++_2M_SDE</li><li>CFG Scale: 3 – 5</li><li>Steps: 20 – 35</li><li>Clip Skip : 2</li></ul>Recommended Prompt:<ul><li>1girl, furry female, anthro, kemono, digitigrade, body fur, rating explicit</li></ul>Recommended Negetive:<ul><li>worst quality, low quality, lowres, off-topic, old, early, signature, watermark, censored, not furry, humanoid face, humanoid feet, humanoid ears, humanoid hands, chibi inset, character doll</li></ul>[DeepViolet]Vpred ModelMake sure you know how to use it properly[DeepBlue]This is a merged model using a different base model instead of the original(V1~V3) model. I tried to keep the feel as similar as possible, but since the model and LoRAs are entirely different, I adjusted the weights one by one during the merge.Recommended Parameters:<ul><li>CFG Scale: 3 – 5</li><li>Steps: 20 – 35</li><li>Sampler: DPM++_2M_SDE</li></ul>Recommended Prompt:<ul><li>1girl, furry female, anthro, kemono, digitigrade, body fur, rating explicit</li></ul>Recommended Negetive:<ul><li>worst quality, low quality, lowres, off-topic, old, early, signature, watermark, logo, english text, censored, not furry, humanoid feet, humanoid pointy ears, humanoid ears, humanoid face, extra ears</li></ul>[Noob AI]Recommand :Sampling method : euler ancestral cfg ppSmapling steps : 24CFG Scales : 1.8Clip Skip : 21girl, furry female, anthro, kemono, digitigrade, body fur, rating_explicit, ~prompt BREAK masterpiece, best quality, amazing quality, very aesthetic, absurdres, newestworst quality, low quality, lowres, off-topic, old, early, chromatic aberration, abstract, jpeg artifacts, scan artifacts, signature, watermark, logo[ILXL]Sampling method : euler aSmapling steps : 28CFG Scales : 6Clip Skip : 2Non furry1girl, rating_explicit, [place], ~prompt masterpiece, best quality, amazing quality, very aesthetic, absurdres, newestFurry1girl, furry female, anthro, kemono, digitigrade, body fur, rating_explicit, masterpiece, best quality, amazing quality, very aesthetic, absurdres, newestNegetive : lowres, (bad), text, error, fewer, extra, missing, worst quality, jpeg artifacts, low quality, watermark, unfinished, displeasing, oldest, early, chromatic aberration, signature, extra digits, artistic error, username, scan, (abstract:0.9), off-topic, 3d, photorealistic, animal, censored, speech_bubble, english_textCuteCrunch [Pony]Sampling method : euler aSmapling steps : 25CFG Scales : 7Clip Skip : 2Prompt : score_9, score_8_up, score_7_up, score_6_up, score_5_up, score_4_up, source_anime, source_furry, furry, female furry, anthro, kemono, solo focus, digitigrade, countershading, snout, ~ promptMol_Keunmix [Pony]Sampling method : euler aSmapling steps : 25CFG Scales : 8.5Clip Skip : 2Prompt : score_8_up, score_7_up, score_6, source_furry, source_anime, ((furry, kemono, anthro)), 1girl, ~ prompt[SD 1.5]YiffyMix + Crosskemo + Etc Models + LoraSampling method : DPM++ 2M KarrasSmapling steps : 40CFG Scales : 9Clip Skip : 2Hires Upscaler : R-ESRGAN 4x+ Anime6BHires Steps : 6Dinoising Strength : 0.4Vea : Furception VAEPrompt : best quality, kemono, anime, female, furry, anthro, full body, solo, Species, two tone body, ~ your prompNegetive : worst quality, low quality, lowers, blurry, unusual anatomy, mutilated, malformed, watermark, amputee, censorship, deformity, sketch, unfinished, baalbuddy, two tails, multiple tailsPrevious description ↓YiffyMix + Crosskemo + Etc Models + LoraSampling method : DPM++ 2M Karras Smapling steps : 20 CFG Scales : 9 Clip Skip : 2 or 1Hires Upscaler : R-ESRGAN 4x+ Anime6BHires Steps : 6Dinoising Strength : 0.45Negetive : worst quality, low quality, lowers, blurry, unusual anatomy, mutilated, malformed, watermark, amputee, censorship, deformity, sketch, unfinished, baalbuddy, two tails, multiple tails, verybadimagenegative_v1.3, bad-hands-5

Mol_Keun Mix (Furry Mix / CuteCrunch)

I worked on removing the noise that used to appear around the eyes and nose in previous versions. As a result, there are quite a few differences from the original DeepBlue model, but I feel that this version understands prompts much better.
You can now use the Euler a sampler without getting any of that noise.


Files with SHA256: de9667c2bdc8bc42572bf1c21aa0a54461dd3e6fda20cfe4dd6913b57b27bb11