Sign In

vibrant slider / saturation slider

2.2k

Updated: Sep 21, 2026

toolvibrantsaturationslider

Recently loaded

Download

1 variant available

bf16 SafeTensor

vibrantslider_anima10.safetensors

BF16, good balance • 4.24 MB

Verified:

Type
LoRA
Stats

929

52

30

Reviews
Published

Sep 1, 2026

Base Model
Training
Steps: 40
Epochs: 1
Usage Tips
Strength: 1
Hash
AutoV2
6778967FC0

The Anima Model is licensed by CircleStone Labs LLC. Copyright CircleStone Labs LLC. IN NO EVENT SHALL CIRCLESTONE LABS LLC BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH USE OF THIS MODEL.

Built on NVIDIA Cosmos

xyz_grid-0087-2942666290.jpg

Try size: 768 x 1024 or 1024 x 1024

Weights: -2.0~2.0

Clip skip: 2

Note: It improves color vibrancy, but unlike typical Anime or SDXL VAE, it causes noticeable interference with details. / 它有改善鮮豔度的效果,但與一般的 Anime 或 SDXL VAE 不同,它對細節的干擾很明顯。(Anima10)

Note: This is a slider LoRA that only affects the color layer, so it is fully universal across ILXL, Pony, and SDXL / 這是slider LoRA,只有針對顏色的那一層去影響,所以對ILXL、Pony跟SDXL的效果都是一樣的,都是通用的。(ILXL / Pony / SDXL)


If you need a VAE with less interference, you can also check the Suggested Resources below. / 如果需要干擾較小的 VAE,也可以查看下面的建議資源。


This img2img colorization workflow is only meant for grayscale sketches, not pure black-and-white line art. If you strictly want to preserve the original drawing, using nano banana 2, grok image 2, or gpt image 2 is highly recommended. (Settings: LoRA weight 3.5, Denoising strength 0.4. Using the Euler sampler will help minimize detail loss/destruction) / 圖生圖上色只能針對灰階稿,不建議黑白稿,如果堅持不破壞原稿,建議採用 nano banana 2、grok image 2 或 gpt image 2。Lora權重 3.5,重繪幅度 0.4,採樣器 Euler 對細節破壞較低。