Sign In

MixTape /// Bossa Nova

Checkpoint Merge
Jun 25, 2023
Base Model
SD 1.5

Beautiful, semi-realistic images with a hint of naughty. It offers a visually pleasing aesthetic and a provocative touch. Bossa Nova brings to life captivating compositions that stimulate the imagination.

Let's create a gallery full of inspiration! Post your masterpieces and let's fill civitai with imagination, creativity, expression and joy!

Just as Bossa Nova enchants with its smooth harmonies and poetic lyrics, the images envelop us in a visually stunning aesthetic, creating an artistic experience that transports us beyond the ordinary.

The Vanguarda version creates illustrations that offer a burst of vibrant color with a smoother, more subtle three-dimensionality.

The Nova version creates illustrations with slightly more realistic proportions and with more details.


Positive Prompt:

(masterpiece, best quality:1.4), (beautiful, perfect, delicate, detailed, intricate, aesthetic:1.2), ...

Negative Prompt:

EasyNegative, BadHands, BadArtist, BadArtistAnime, BadPrompt, BadImages, (low quality, worst quality:1.4), (monochrome, grayscale, sketch:1.2), (logo, signature, text, words, username, artist name, watermark), (loli:1.2), ...

Embeddings (Textual Inversion): EasyNegative, BadHands, BadArtist, BadArtistAnime, BadPrompt, BadImages

Sampling method: DPM++ SDE Karras

Sampling steps: 30 ~ 50

Hires. fix:

Sampler: 4x-UltraSharp

Hires steps: 6 ~ 10

Denoising strength: 0.5

Upscale by: 2 (resize from 512x768 to 1024x1536)

Clip skip: 2

CFG Scale: 7 ~ 8

VAE: kl-f8-anime2 VAE

Merge Recipe

Merged models are the following:

Vanguarda Steps:

  1. BaseModel = K-main x 0.3 + CuriousMerge 2.5D x 0.7

  2. MixTape = BaseModel x 0.5 + DucHaitenNiji x 0.5

Nova Steps:

  1. BaseModel = K-main x 0.7 + CuriousMerge 2.5D x 0.3

  2. MixTape = BaseModel x 0.5 + DucHaitenNiji x 0.5


All used models here have "Have different permissions when sharing merges" license permission, so I do not try to add any restriction on it.

CreativeML-Open Rail-M

This model permits users to:

✔Use the model without crediting the creator

✔Sell images they generate

✔Run on services that generate images for money

✔Share merges using this model

✔Sell this model or merges using this model

✔Have different permissions when sharing merges