Stable Diffusion (Best Models) p.4

Wassimgouia
3 min readAug 21, 2023

--

Before moving on to another important subject of Stable Diffusion Make sure to see the previous blog posts

Stable Diffusion (Basics) p.1:

Stable Diffusion (Prompting and Beyond) p.2 :

Stable Diffusion (Beyond Prompting) p.3 :

In this post, we will be discussing the best models for Stable Diffusion in both versions of SD1.5 and SDXL. But before we dive into that, make sure to check out the previous blog posts on Stable Diffusion basics, prompting, and beyond. We’ll be building on those concepts in this post.

Let’s begin by revisiting the image that was used in the previous blog post.

This image was generated using the Model :

dreamshaper_6BakedVae.safetensors

And here I will show you again the prompt and the other proprieties of this image generation

Prompt:

a goddess of Ireland, Beautiful face, clear skin, blue eyes, a cute little smile, red hair, Wearing green Clothes, and a green hat in her head, Portrait mode, highly detailed clothes, 8k

Negative Prompt:

ugly, tiling, poorly drawn hands, poorly drawn feet, poorly drawn face, out of frame, extra limbs, disfigured, deformed, body out of frame, bad anatomy, watermark, signature, cut off, low contrast, underexposed, overexposed, bad art, beginner, amateur, distorted face, blurry, draft, grainy

And Finally:

SD1.5:

After some searches I’ve come up with the best models for the SD1.5:

For Photo-Realism:

majicmixRealistic_betterV2V25.safetensors :

photon_vl.safetensors :

chilloutmix_NiPrunedFp32Fix.safetensors :

beautifulRealistic_v60.safetensors :

absolutereality_v16.safetensors :

For Illustration/Digital-Paintings/Anime:

deliberate_v2.safetensors :

Protogen_V2.2.ckpt :

ProtogenV22Anime_protogenV22.safetensors :

revAnimated_v122.safetensors :

SDXL:(The SDXL models basically works with every style so i made the styles in between)

sd_xl_base_1.0_0.9vae.safetensors :

sd_xl_refiner_1.0.safetensors :

dreamshaperXL10_alpha2X110.safetensors :

rundiffusionXL_beta.safetensors :

ambienceSDnaLsafetensors :

Conclusion:

There are undoubtedly numerous excellent models available, but after extensive experimentation, I’ve found that these particular ones align best with a specific style in my personal experience.

follow My X :

--

--