If you’re building a sci‑fi world, this LoRA is your trusted companion for setting the right mood. Whether you want a post‑apocalyptic future or a high‑tech dystopia, it delivers that unique sci‑fi atmosphere. Just remember: it enhances your prompt, it doesn’t replace it. 😊
This LoRA delivers a distinctive sci‑fi style while keeping both the form and fine details intact where it matters. It highlights the best of cyborgs, androids, and robots: mechanical precision paired with bold, impactful visuals. The characters’ gazes are intense and unsettling, and their very presence is designed to intimidate.
Architecture and surroundings function as a key visual character, giving the scene history and scale.
Lighting isn’t meant to be bright—it’s meant to build atmosphere. It casts deep shadows where anything could be hiding (and most likely is 😉). Every highlight and every dark corner works to create mood, not just to show what’s there.
The LoRA strikes a balance between realism and stylization: enough detail to make the world feel believable, and enough mood to stir a sense of unease.
Perfect for moody concept art, stylized realism, dystopian surroundings, and tense scenes where the characters and environment together tell the story with a distinct sci‑fi eerie vibe.
Check the attached images to get an idea of this LoRA’s capabilities. Drop your results here — I want to see them all! And have fun!
Here are a few words about the settings I prefer for this LoRA (as a Forge Neo / SwarmUI user):
Strength: 0.9 (tested working range: 0.7–1.0)
Sampling Method / Schedule Type:
2-stage generation: use any sampler/scheduler combo you prefer (cfg for both stages: not less than 1.1). The denoising value and number of steps depend on your chosen sampler/scheduler combo — I switch between different combos depending on the specific image, so there's no single "correct" preset. Pick what works best for your scene.
Upscaler: Remacri
And yes, I do use negative prompts — check the attached images for an example.
Description
V2 isn’t just ‘V1 but better’ — it’s a whole new flavor, much like the two versions of my KreaDream LoRA. I used different training settings, swapped in about half a new dataset, and slightly increased the number of images. As always, all images in the dataset are top‑tier, hi‑res.
The result is a smaller LoRA file with a darker, more intense sci‑fi style—delivering extra punch in the details: cybernetics, robotic/engineering elements, and that gritty futuristic feel.
So use V2 if you’re looking for this kind of touch. Check the attached images to get a feel for what V2 can do.
FAQ
Comments (4)
The second version of your LoRA is very different from the first one, and I am very happy about it. Now it's really a fantastic style.And it does not replace the first version, but complements it.
Thank you so much for the feedback! 🫶 So glad you’ve gotten a feel for both versions. Indeed, the idea behind them was to bring out different flavours of the core sci-fi aesthetic.
From the moment I understood the weakness of my flesh, it disgusted me. I craved the strength and certainty of steel. I aspired to the purity of the Blessed Machine. Your kind cling to your flesh, as though it will not decay and fail you. One day the crude biomass you call the temple will wither, and you will beg my kind to save you. But I am already saved, for the Machine is immortal… Even in death I serve the Omnissiah.
You speak the truth as the cogitators know it. This LoRA holds secrets whispered by machine-spirits long forgotten. It does not merely generate. It interprets the will of the Omnissiah through the language of gear and data. Thank you for hearing the hidden chorus. Let us forge images worthy of the Machine God’s gaze.