Another test lora for Anima!

Since it is test version I need your feedback very much to understand is Anima really SDXL killer or not.

I updated dataset and fixed many problems so I'll post NoobAI / Epsilon / Illustrious version on this page too soon!

Character tags:

Jaina Proudmoore, 1girl,

white hair, blonde hair, streaked hair, single braid, blue eyes, anchor necklace,

Lord Admiral outfit:

single pauldron, cape, cropped jacket, corset, detached sleeves, gloves, blue pants, long skirt, single gauntlet, knee boots, cleavage, belt pouch, holding staff, planted

My original hot outfit for her:

detached sleeves, black gloves, corset, bare shoulders, fishnet thighhighs, thigh boots, miniskirt, blue skirt, cleavage, skindentation

For hot Jaina:

nipple piercing, nipple rings, anchor piercing

horde tattoo, pubic tattoo (or ass tattoo, breast tattoo, it works on Anima without extra lora).

New! Style control:

3D - for classic r34 Jaina.

3D, realistic - for modern 3D realistic model version of Jaina.

For interactions:

orc, zandalari troll, troll /(warcraft/);

Garrosh Hellscream, orange skin, dark skin, facial hair, sharp teeth, bald, muscular male, interspecies, brown eyes, topless, brown pants, chest tattoo

night elf /(warcraft/), Tyrande Whisperwind;

dark elf, Sylvanas windrunner;

tauren, worgen, furry with non-furry

Also Anima supports natural language because of LLM adapter + qwen text encoder! If scenes are difficult try to use sentences for description!

SETTINGS FOR ANIMA PREV 3:

I'll do article about Anima model according to my 2 weeks tests SOON. Need little more tests about samplers and schedulers.

Short guide for now:

Same aesthetic settings (masterpiece, best quality, good quality, highres, absurdres, worst quality, low quality, bad quality, lowres, extra digits in negative.) as for V-pred but you don't need "very awa" in positive and "worst aesthetic" in negative because these tags were exclusively for NoobAI V-pred.

Also don't use "score_". Anima supports these tags but they make you pics AI-slop like Pony because score-based aesthetic tags can cause overcontrast.

Generations settings:

Euler A / Euler for txt2img, Euler for highres fix (euler a can break backgrounds in highres fix stage so I prefer euler highres fix. If you don't have option to choose in forge / forge neo go to settings and choose "[Hires. fix]: Show checkpoint, sampler, scheduler, and cfg options"

CFG 3. You can try CFG 4-5 but personally I like visual of CFG 3 more especially for 3D style pics.

30-40 steps because Anima is more gluttonous than SDXL and 10-20 highres steps, 0.25 - 0.4 denoising for highres (depends on scene, sometimes txt2img pic is already masterpiece, lol). Adetailer - use very high pixel padding - about 176-232 pixels instead of 32 pixels because Anima understands context much better. Denoising of adetailer - 0.25-0.3.

My preview can help!

Post you works in gallery and your opinion in comments! I need your feedback. 😊