OstAnderMix版本v1.0 (ID: 606755)

私は純粋なアニメ絵も好きですが、ややリアル調のイラスト ｰ 「2.25D」とでも言えばいいでしょうか ｰ がより好きです,このモデルはそんな自分好みのイラストを生成するにはどうすればよいか、試行錯誤しながらマージしたものです,

I like purely animated pictures, but I prefer slightly more realistic illustrations - "2.25D", if you like. This model was merged through trial and error to see how I could generate illustrations to my liking.

---

License:

Fair AI Public License 1.0-SD

• You should share the merge recipe if you release a model merged with OstAnderMix

• This license does not add any restrictions on your generated images.

• For more details, please see the license section of ANIMAGINE XL 3.1,

---

描述:

V.1.0.0 (2024.06.29)

Checkpoints:

• Animagine XL V3.1

• anima-Pencil-XL v4.0.0

• AniEase v2.1

• AniEase v2.4

• Anime from HaDeS XL v2.1

• Async's MIX XL v3.1

• Async's MIX XL v3.2

• Better than words v3.0

• blue_Pencil-XL v7.0.0

• Cherry Picker XL v3.0

• Dark Pizza XL Origin 大个披萨XL 原味儿 v4.0

• ICBINP XL v2.0

• ICBINP XL v6

• LEOSAM's HelloWorld XL 3.2DPO

• LEOSAM's HelloWorld XL 7.0

• SD XL v1.0 VAE fix

• VoidnoiseCoreXL R1486

Recipe

1. anima-Pencil-XL v4.0.0+ (Async's MIX XL v3.1 - Animagine XL V3.1) x 1.0, Add difference(trainDifference)

2. [1] + (Async's MIX XL v3.2 - Async's MIX XL v3.1) x 1.0, Add difference(trainDifference)

3. [2] + (AniEase v2.4 - AniEase v2.1) x 0.3, Add difference(trainDifference)

4. Cherry Picker XL v3.0 + LEOSAM's HelloWorld XL 7.0 -LEOSAM's HelloWorld XL 3.2DPO) x 1.0, Add difference(trainDifference)

5. [4] + (ICBINP XL v6 - ICBINP XL v2.0) x 1.0, Add difference(trainDifference)

6. ([5].fp16 x 0.6 +Better than words v3.0 x 0.4) x 0.7 + VoidnoiseCoreXL R1486 x 0.3(sum Twice)

7. Async's MIX XL v3.2 + [6]
   base_alpha:0, weight:(0.0,1.0,0.9166666667,0.8333333333,0.75,0.6666666667,0.5833333333,0.5,0.4166666667,0.3333333333,0.0,0.0833333333,0.1666666667,0.25,0.3333333333,0.4166666667,0.5,0.5833333333,0.6666666667,0.75,0) [GRAD_V]

8. Animagine XL V3.1 + SD XL v1.0 VAE fix
   base_alpha:0, weight:(0.0,1.0,0.9166666667,0.8333333333,0.75,0.6666666667,0.5833333333,0.5,0.4166666667,0.3333333333,0.0,0.0833333333,0.1666666667,0.25,0.3333333333,0.4166666667,0.5,0.5833333333,0.6666666667,0.75,0) [GRAD_V]

9. [8] x 0.75+ darkPizzaMix x0.25, Weight sum(normal)

10. [7] x 0.7 + [9] x 0.3, Weight sum(cosineA)

11. blue_Pencil-XL v7.0.0 + Anime from HaDeS XL v2.1
    base_alpha:0, weight:(0.0,1.0,0.984809027777778,0.940972222222222,0.87109375,0.777777777777778,0.663628472222222,0.53125,0.383246527777778,0.222222222222222,0.5,0.687065972222222,0.871527777777778,0.94921875,0.777777777777778,0.616753472222222,0.46875,0.336371527777777,0.222222222222222,0.12890625,0) [R_SMOOTHSTEP*3]

12. [11] x 0.25 + [3] x 0.75,Weight sum(tensor2)

13. [10] + [13]
    base_alpha:0, weight:(0.0,0.0,0.0151909722222222,0.0590277777777778,0.12890625,0.222222222222222,0.336371527777778,0.46875,0.616753472222222,0.777777777777778,0.5,0.312934027777778,0.128472222222222,0.0507812500000004,0.222222222222222,0.383246527777778,0.53125,0.663628472222223,0.777777777777778,0.87109375,0) [SMOOTHSTEP*3]

14. Removed junk keys from models in [13]

训练词语: