circlestone-labs/Anima · Discussions

#135 opened 1 day ago by

ArranEye

is `~` being tokenized to unk?

#134 opened 2 days ago by

sorryhyun

Turbo Lora

#133 opened 2 days ago by

kdutt2000

Is there any approximate date for the full release? Also, a few questions about prompting.

#132 opened 2 days ago by

Doramu0917

Request for ControlNet Support

👍 2

#131 opened 3 days ago by

lastxuam

mikkoukun style and Characters

#130 opened 6 days ago by

have Lora training times improved? and what am i missing...

14

#129 opened 6 days ago by

aubreyz

artifacts

#127 opened 8 days ago by

What model was used to generate natural language captions for the dataset?

#126 opened 9 days ago by

analogspiderweb

Model output diversity drop

4

#125 opened 10 days ago by

LLLMI

Inpainting workflow?

#124 opened 10 days ago by

JonathanZ

Is there a Discord server dedicated to Anima related stuff? (training, captioning, prompting, workflow sharing, etc.)

#123 opened 11 days ago by

alright-bibi

Proposal: Modulation Guidance — making AdaLN text-aware for quality steering

👍❤️ 15

#122 opened 13 days ago by

sorryhyun

Hi ;P

👀🔥 3

#121 opened 14 days ago by

Will prompt adherence include seperate character tagging

4

#120 opened 14 days ago by

Nognig

Help with Character LoRA Training for Anima Preview2 – Poor Generalization, Only Frontal ID-Style Shot Matches Target Character

🤯 1

#119 opened 14 days ago by

brendonlpbesler

Full Finetuning Questions

👍 12

#118 opened 15 days ago by

motimalu

A small suggestion: I think this model should get a new name

#117 opened 15 days ago by

sheldon88

character tags

#116 opened 15 days ago by

vekitan

Any news about the final version of the model?

4

#115 opened 16 days ago by

JarvisS666

Hi Just a tip For the final version

🤯🧠 6

#114 opened 17 days ago by

Anime training

#113 opened 17 days ago by

Artist style becomes inconsistent with multiple artists / longer prompts — detailed analysis & suggestions

🧠👍 19

25

#112 opened 17 days ago by

Henry-Yan

When using 1-boy pants as a prompt, women's leggings will dominate in image generation.

#111 opened 19 days ago by

Namame

'conv_in.weight'

#110 opened 20 days ago by

CaptClueless

Can't upscale a batch of multiple images through refiner pipeline?

#109 opened 20 days ago by

synta

regarding Anima training

👍 7

#108 opened 21 days ago by

siegmundwulfe

Cosmos-Predict2.5?

#107 opened 21 days ago by

QRTN

Lora training parameters

#106 opened 21 days ago by

sw0ad

Getting censor bars with certain artists, even with censored in negative prompt.

#104 opened 22 days ago by

Euchale

Impact of quility tags on lora training

#103 opened 24 days ago by

KeMiliUs

stabilise images

#102 opened 25 days ago by

tygu

Questions about the input

7

#101 opened 28 days ago by

Aetherlyn

Training question from a beginner

#100 opened 28 days ago by

bnotoe

Inconsistencies with directional prompts in the Anima model

#99 opened 28 days ago by

daijava

Clip type? Stable diffussion, cosmos or qwen?

#98 opened about 1 month ago by

OurpleSlop

Plans for an Edit Version of the model?

#97 opened about 1 month ago by

Mudflap

Strategy for captioning for Anima and captioning tools

➕ 3

5

#96 opened about 1 month ago by

yokoo

Watermark appearing despite negative prompt tag

👀 1

#94 opened about 1 month ago by

Bloshechka

Is there a consistent way to prompt multiple characters together?

#93 opened about 1 month ago by

CulturedDiffusion

Training LyCORIS for Anima

#92 opened about 1 month ago by

njs45

Bad results????? Worse than SDXL finetunes?

😔🤯 4

#91 opened about 1 month ago by

UberMao

clarification please

5

#90 opened about 1 month ago by

soupcheese

Hiiiiii

👍 8

#88 opened about 1 month ago by

Ysshisusgsh

Suggestion: Using Karpathy’s "AutoResearch" for Anima training optimization

🤯😔 4

5

#87 opened about 1 month ago by

kdutt2000

Corrupted data in dataset?

#86 opened about 1 month ago by

LLLMI

Escaping brackets with backslashes in captions

#85 opened about 1 month ago by

degurshaft

Diffusers?

#84 opened about 1 month ago by

little-lake-studios

Potential selection bias in the dataset scoring pipeline (ai-generated tag behavior)

👍 10