Sign In

HydraMix

132
1.1k
17
Verified:
SafeTensor
Type
Checkpoint Trained
Stats
1,133
Reviews
Published
Jun 13, 2023
Base Model
SD 1.5
Trigger Words
aerith
bowsette
mercy
widowmaker
tracer
dva
black cat dva
star guardian jinx
star guardian lux
irelia
+13 more
Hash
AutoV2
772D12793D
default creator card background decoration
SDXL Training Contest Participant
wier's Avatar
wier

Trying to find a balance between my other models, this one is capable of great realistic fantasy portraits, while also being able to merge different concepts, hybrids, humanoid creatures. Capable of nsfw results too, use nsfw naked nude in negative to avoid them.

Works kinda well with 896x1152 native resolution but starts to lose coherency.

Main models used for merging (other than Wyvern, Succubus, Chimera):

AIOMonserGirlsV3, AmIRealV3, AnimalHumanHybridsV1, BB95FurryMixV8, PirsusEpicRealism(V1.3,1.5,2.0), SpeciesV0.8, YiffyMix2.2
+ many more in smaller parts

+ Additional training with animals, gen1 pokemons, instagram pics, reddit photography, some lol and ow characters (from datasets shared here), celebs at different ages (NAME, Xyo, Y years old)

More realistic by default.

Examples at 768x960, DPM++ 2M Karras, CFG 7, 20 steps

Example 1:

Prompt: name

Example 2:

Prompt: name

Negative prompt: child cartoon rendering 3d doll drawing painting

Example 3:

Prompt: anime version of name

Negative prompt: child