Guizmus/Experiments

Guizmus' test models

Those are medium to low quality models, single focus usualy, that can bring some fun but not really do significant things.

I'm not sharing the diffusers, just the ckpt and some examples.

You can also find my models relating to the Picture of the Week on Stable Diffusion discord here usion discord here

---

SDArt v2 (13 PoW)

* training: 593 pictures, 59 300 steps on 1.5e-6 polynomial LR in EveryDream2, base 1.5 +VAE * Activation token : none, trained as artistic model. Themes trained : * lots of medium and generic concepts : anime artwork, painting, artwork, 3D render, realistic photo, portrait, close up, ... * PoW Alice in Diffusionland : Alice, cat, rabbit, hatter, psychedelic * PoW Catstravaganza : Cat. in lots of situations, with clothes, ... * PoW Cosmic Horrors : Horror, tentacles, mist, black and white, cows, cthulhu, glowing eyes * PoW Abstract Realities : Broken, abstract, colorful * PoW Elemental Flowers : ice, fire, water, earth, flowers, elements * PoW Futuristic Fashion : futuristic clothes, tech, neon lights, cyberpunk * PoW Underwater Worlds : underwater, mermaid, coral, sea * Also included but auto captionned, there is less focus on specific tokens like the previous ones : Imaginary friend, Never ending loop, someone, something, somewhere, synesthesia. * Model link * token list * dataset link

SDArt mix v1 (7 PoW)

• training: 329 pictures, 32 900 steps on 1.5e-6 polynomial LR in EveryDream2, base 1.5 +VAE
• Activation token : SDArt also look at the token list pdf
• Model link
• token list
• dataset link

---

Bad Cosplay man v1

• training: 42 pictures, 24 000 steps on 1e-6 polynomial LR in EveryDream
• Activation token : Bad cosplay man as
• Model link
• dataset link

---

Rick Roll v1

• training: 20 pictures, 2400 steps on 1e-6 polynomial LR with 8 bit adam
• Activation token : Rick Roll
• Model link

---

TMNT v1.0

• training: 11 pictures, 1000 steps on 5e-6 polynomial LR with 8 bit adam
• Activation token : TMNT person
• Model link

---

Sid v1.0

• training: 10 pictures, 2500 steps on 1e-6 polynomial LR
• Activation token : Sid
• Model link

---

Kirby v1.0

• training: 10 pictures, 2500 steps on 1e-6 polynomial LR
• Activation token : Kirby
• Model link

---

Simsbooth v1.0

• training: 122 pictures, 18k steps on 1e-6 polynomial LR
• Activation token : Simsbooth person
• Model link

---

Hogleg Style v2.4

• training: 5k steps lr 1e-6, 122 samples, then 2k more steps on a lower lr and a rebalanced dataset
• Activation token : HogLeg Style
• Model link

---

Illusion v1.0

• training: 1k2 steps, 12 samples
• Activation token : Illusion drawing
• Model link

---

NikeAir v1.0

• training: 1k2 steps, 10 samples
• Activation token : NikeAir Shoes
• Model link