konapieces commited on
Commit
f47f5ad
1 Parent(s): 3ee64c2

Upload README.md

Browse files
Files changed (1) hide show
  1. README.md +75 -0
README.md CHANGED
@@ -171,6 +171,8 @@ lastpiece.vae.contrast (konapieces) - https://huggingface.co/konapieces/Lastpiec
171
  <summary>VoidnoiseCore_R1311</summary>
172
  <div>
173
 
 
 
174
  # 本モデルの概要 (Overview of this model)
175
 
176
  本モデルは<a href="https://huggingface.co/spaces/CompVis/stable-diffusion-license" target="_blank">CreativeML OpenRAIL-M</a>ライセンスの訓練モデル、SD1.5追加学習モデル、またそれらのみを使用したマージモデルで製作したモデルになります。<br>
@@ -350,6 +352,79 @@ BREAK
350
  <img src="https://huggingface.co/konapieces/VoidnoiseCore/resolve/main/images/F1725/sample2.png" width="512">
351
 
352
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
353
  </div>
354
  </details>
355
 
 
171
  <summary>VoidnoiseCore_R1311</summary>
172
  <div>
173
 
174
+ ![](images/R1311/logo.png)
175
+
176
  # 本モデルの概要 (Overview of this model)
177
 
178
  本モデルは<a href="https://huggingface.co/spaces/CompVis/stable-diffusion-license" target="_blank">CreativeML OpenRAIL-M</a>ライセンスの訓練モデル、SD1.5追加学習モデル、またそれらのみを使用したマージモデルで製作したモデルになります。<br>
 
352
  <img src="https://huggingface.co/konapieces/VoidnoiseCore/resolve/main/images/F1725/sample2.png" width="512">
353
 
354
 
355
+ </div>
356
+ </details>
357
+
358
+ <details>
359
+ <summary>VoidnoiseCore_R1733</summary>
360
+ <div>
361
+
362
+ ![](images/R1733/logo.png)
363
+
364
+ # ▼ 本モデルの概要 (Overview of this model)
365
+
366
+ 本モデルは<a href="https://huggingface.co/spaces/CompVis/stable-diffusion-license" target="_blank">CreativeML OpenRAIL-M</a>ライセンスの訓練モデル、SD1.5追加学習モデルのみを使用したマージモデルになります。<br>
367
+ VoidnoiseCore R1733は、前作のVoidnoiseCore R1311をベースに作成されています。<br>
368
+ LoRAやLyCORISを使用しなくても、目力が強い人物を描写することに特化しており、フォトグラフィ調の構図を得意としています。<br>
369
+ 今回より、アジア女性に限らず、欧米女性などの描写も美しく描写することができる、前作の完全上位互換を目指して作成しています。<br>
370
+ <br>
371
+ This model will be a merged model, using only the training model and SDXL1.0 additional learning model from the <a href="https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0/blob/main/LICENSE.md" target="_blank">CreativeML OpenRAIL++-M</a> license.<br>
372
+ VoidnoiseCore R1733 is based on the previous VoidnoiseCore R1311.<br>
373
+ It specializes in depicting people with strong eyes without using LoRA or LyCORIS, and excels in photographic composition.<br>
374
+ From this time onward, we are creating a fully upward compatible version of our previous work that can beautifully depict not only Asian women but also Western women and other types of women.<br>
375
+
376
+
377
+ # ▼ 推奨設定 (Recommended settings)
378
+
379
+ - Sampler: Restart
380
+ - Step: 35 - 40
381
+ - CFG scale: 11.0 - 12.0
382
+ - Denoising strength: 0.55 - 0.65
383
+ - Clip skip: 2
384
+ - Hires upscale: 2
385
+ - Hires steps: 15 - 20
386
+ - Hires upscaler: R-ESRGAN 4x+
387
+ - ENSD (Eta noise seed delta): 31337
388
+ - VAE: <a href="https://huggingface.co/konapieces/VoidnoiseCore/blob/main/models/R0829/vae/voidnoise.vae.tuned.safetensors" target="_blank">voidnoise.vae.tuned</a>
389
+ - Embeddings: <a href="https://civitai.com/models/56519/negativehand-negative-embedding" target="_blank">negative_hand</a>
390
+ - Extensions : FreeU, Tiled VAE, Tiled Diffusion
391
+
392
+ # ▼ 出力サンプル (Sample)
393
+
394
+ [Prompt]<br>
395
+
396
+ ```
397
+ (realistic, photo-realistic:1.37), (masterpiece), (best quality:1.4), (ultra high res:1.2),(RAW photo:1.2), (sharp focus:1.3), (face focus:1.2), vibrant details, hyperrealistic, elegant,
398
+ kawaii,cute,(1 woman, solo),inverted bob, bangsamethyst hair,azure eyes,drooping eyes,longeyelashes,eyeshadow,eyeliner,(smile),((hipster, loose turtleneck sweater, overcoat):1.37),large breasts,
399
+ BREAK
400
+ park,fallin snow, simple background, ((night,bokeh):1.37)
401
+ ```
402
+
403
+ [Negative prompt]<br>
404
+ ```
405
+ EasyNegativeV2,negative_hand-neg,(worst quality, bad quality:1.4), (extra fingers, deformed hands, polydactyl:1.5), (bad hands, bad fingers, bad arm, missing finger:1.5), text, nsfw
406
+ ```
407
+
408
+ <img src="https://huggingface.co/konapieces/VoidnoiseCore/resolve/main/images/R1733/sample1.png" width="512">
409
+
410
+ [Prompt]<br>
411
+
412
+ ```
413
+ (realistic, photo-realistic:1.37), (masterpiece), (best quality:1.4), (ultra high res:1.2),(RAW photo:1.2), (sharp focus:1.3), (face focus:1.2), vibrant details, hyperrealistic, elegant,
414
+ kawaii,cute,(1 woman, solo),french twist, bangsamethyst hair,azure eyes,drooping eyes,longeyelashes,eyeshadow,eyeliner,earrings,(smile),
415
+ ((A polka dot swing dress with a sweetheart neckline):1.37),
416
+ BREAK
417
+ simple background, ((night,bokeh):1.37),
418
+ ```
419
+
420
+ [Negative prompt]<br>
421
+ ```
422
+ EasyNegativeV2,negative_hand-neg,(worst quality, bad quality:1.4), (extra fingers, deformed hands, polydactyl:1.5), (bad hands, bad fingers, bad arm, missing finger:1.5), text, nsfw
423
+ ```
424
+
425
+ <img src="https://huggingface.co/konapieces/VoidnoiseCore/resolve/main/images/R1733/sample2.png" width="512">
426
+
427
+
428
  </div>
429
  </details>
430