A new merge enters the stage ~ Keij, specially created for the Vidu x Civitai Contest, shaped by my style: a semirealistic fusion of reality and fantasy, rendered with emotion, motion, and visual storytelling in mind.
Keij is not about technical perfection or heavy quality prompting ~ it’s about capturing ideas and moments. These are not storybook images, but cinematic snapshots ~ ideal for image-to-video workflows, breathing with atmosphere and depth.
🎵 A vivid mix of movement, emotions, color, light, clothing, and fine details ~ expressive and full of energy
🌸 Perfectly suited for 864×1536 or 720×1280 (portrait format) ~ other formats work just as well
🎵 Works beautifully without extensive quality prompts ~ often "absurdres" is all u need
🌸 Designed for dynamic effects ~ even with minimal prompting, thanks to focused merging
🎵 Adetailer is recommended for distant shots ~ for close-ups and portraits, it’s up to ur taste
🌸 Recommended sampler: • Euler a ~ soft, more fluid look
~ the merge was specifically created with Euler a in mind, as it reflects my personal aesthetic preference
~ by using other sampler like DPM++ 2M: For lower resolutions (like 720×1280), I recommend increasing the number of steps, as the model contains a lot of fine detail. More steps help retain a smoother, more coherent appearance
🎵 CFG: 5+
🌸 Steps: 30+
🎵 Negative prompt: optional ~ “lowres, bad anatomy” works when needed
Even though Keij is "just a merge", it’s crafted with care ~ carrying my own visual voice and merging philosophy. Each generation is a step deeper into a vibrant universe ~ one that doesn’t need to be over-described to shine.
Let Keij open the gate to ur own dreamlike realities.
🌸 I’m truly looking forward to seeing ur creations with Keij.
If u enjoy this style of visual poetry and the way it tells stories through light, color, and emotion, a like means a lot and lets me know it's resonating with u.
Let’s keep the gates to creativity wide open ~ and see what worlds we can build together. 🎵✨
best regards ~♥ Reij
recommended onsite gen settings:
+prompt: absurdres
-prompt: bad anatomy, lowres,
Euler a
CFG 5
steps 40-50
hires settings: Denoise 0,30
Keij is a highly specialized Image generation AI Model of type Safetensors / Checkpoint AI Model created by AI community user reijlita. Derived from the powerful Stable Diffusion (Illustrious) model, Keij has undergone an extensive fine-tuning process, leveraging the power of a dataset consisting of images generated by other AI models or user-contributed data. This fine-tuning process ensures that Keij is capable of generating images that are highly relevant to the specific use-cases it was designed for, such as base model.
With a rating of 0 and over 0 ratings, Keij is a popular choice among users for generating high-quality images from text prompts.
Yes! You can download the latest version of Keij from here.
To use Keij, download the model checkpoint file and set up an UI for running Stable Diffusion models (for example, AUTOMATIC1111). Then, provide the model with a detailed text prompt to generate an image. Experiment with different prompts and settings to achieve the desired results. If this sounds a bit complicated, check out our initial guide to Stable Diffusion – it might be of help. And if you really want to dive deep into AI image generation and understand how set up AUTOMATIC1111 to use Safetensors / Checkpoint AI Models like Keij, check out our crash course in AI image generation.
VAE baked in
merge reciepe based on my own merges:
• including Illustrij15, Feijon (which is based on Iluvorija including unpublished Lora) and Edition K:POP beta
Go ahead and upload yours!
Your query returned no results – please try removing some filters or trying a different term.