models/AnyLoRA - Checkpoint - bakedVae ftMse fp16 NOT-PRUNED

AnyLoRA - Checkpoint - bakedVae ftMse fp16 NOT-PRUNED

|
12/21/2025
|
5:55:51 AM
| Discussion|
0
Hutao anime figure sitting with a smile, dressed in a detailed black outfit with gold accents and red flower hair accessory.

Recommended Prompts

1girl, looking at viewer, solo, upper body

Recommended Negative Prompts

(painting by bad-artist-anime:0.9), (painting by bad-artist:0.9), error, text, blurry, cropped, worst quality, watermark, jpeg artifacts, low quality, normal quality, signature, username, artist name, (worst quality, low quality:1.4), bad anatomy

Recommended Parameters

samplers

DPM++ SDE Karras, Euler a, DPM++ 2M Karras

steps

20 - 30

cfg

8 - 9

resolution

512x704, 512x768

vae

good vae, baked vae versions

other models

AnyLoRA_noVae_fp16-pruned (b0c91a8a93), AnyLoRA_bakedVae_fp16_NOTpruned (b775b13509)

Recommended Hires (High-Resolution) Parameters

upscaler

R-ESRGAN General 4xV3, R-ESRGAN 4x+, 4x_foolhardy_Remacri, R-ESRGAN AnimeVideo

upscale

1.7 - 1.8

steps

10 - 20

denoising strength

0.28 - 0.45

Tips

Use the pruned version for training to reduce VRAM usage.

Train with CLIP skip 2 and booru style tags for optimal style accuracy.

Adjust the weight to around 0.65 and offset to 1 for best results (e.g., via ComfyUI).

Creator Sponsors

Add a ❤️ to receive future updates.

Do you like what I do? Consider supporting me on Patreon 🅿️ or feel free to buy me a coffee

Available on the following websites with GPU acceleration:

AnyLoRA

Add a ❤️ to receive future updates.
Do you like what I do? Consider supporting me on Patreon 🅿️ or feel free to buy me a coffee

For LCM read the version description.

Available on the following websites with GPU acceleration:

Remember to use the pruned version when training (less vram needed).
Also this is mostly for training on anime, drawings and cartoon.

I made this model to ensure my future LoRA training is compatible with newer models, plus to get a model with a style neutral enough to get accurate styles with any style LoRA. Training on this model is much more effective compared to NAI, so at the end you might want to adjust the weight or offset (I suspect that's because NAI is now much diluted in newer models). I usually find good results at 0.65 weigth that I later offset to 1 (very easy to do with ComfyUI).

This is good for inference (again, especially with styles) even if I made it mainly for training. It ended up being super good for generating pics and it's now my go-to anime model. It also eats very little vram.

Get the pruned versions for training, as they consume less VRAM.

Make sure you use CLIP skip 2 and booru style tags when training.

Remember to use a good vae when generating, or images wil look desaturated. Or just use the baked vae versions.

Contributor

Previous
Hu Tao Genshin Impact | Character Lora 2293 - v1.0
Next
Dark Futuristic Circuit Boards - Dark Futuristic Circuit Boards

Model Details

Model type

Checkpoint

Base model

SD 1.5

Model version

bakedVae ftMse fp16 NOT-PRUNED

Model hash

8a952cafe9

Creator

Discussion

Please log in to leave a comment.