Downloads · 30 days
60
33% of all-time downloads
bsflash/zIT_Arthouse_Cinema
zIT_Arthouse_Cinema is a text-to-image model from bsflash. Use it when you need an image from a text prompt. It is set up for diffusers.
This LoRA was trained for Z-Image Turbo to reproduce a contemporary arthouse cinema look with an indie aesthetic. It adds subtle realistic film-stock characteristics such as grain, soft bloom, highlight halation, gent…
Downloads · 30 days
60
33% of all-time downloads
All-time downloads
184
Public
Repo size
172 MB
Likes
1
Public
Click a slice to open those files.
.safetensors170 MB · 99%
From the Hugging Face model README
This LoRA was trained for Z-Image Turbo to reproduce a contemporary arthouse cinema look with an indie aesthetic. It adds subtle realistic film-stock characteristics such as grain, soft bloom, highlight halation, gentle vignetting, depth of field, and professional color grading for a cinematic finish.
Because it is based on independent film references, it may also influence character appearance and facial styling toward a more natural, everyday look rather than a polished blockbuster-style beauty standard. In some generations, results may appear slightly too dark, so minor exposure or brightness adjustments in your prompt may help.
For best results, generate images in landscape 16:9 format with the trigger word: bsArthouse, and the LoRA weight strength between 0.6 and 0.7.
While this LoRA works with simple scene descriptions in natural language, it performs better with structured prompts that include film-production keywords. Compare the example images to see the difference in prompting style and output.
Example images:

If your prompts don’t produce the desired results, try this structured template:
SCENE:
<Your scene description>
COMPOSITION:
Dynamic off-axis framing with extreme negative environmental space and an angled perspective to accentuate depth and create visual tension. Sculpted light with negative fill.
STYLE:
A cinematic live-action still shot on an ARRI Alexa 65 with an 80mm prime lens. An A24-inspired indie aesthetic grounded in photographic realism, practical backlighting, and professional digital-to-film color grading. Shallow selective focus with subtle blur, soft highlight bloom, gentle vignetting, highlight halation, and tactile texture.
Prompt Enhancement
You can also use the included bsFlash_Arthouse_Prompt_Enhancer.zip file to generate structured prompts with your preferred LLM. A vision model can also generate structured prompts from a reference image.
Gallery
<Gallery />All example images were generated using: Draw Things.
The official version of this LoRA is only available on Hugging Face. Please do not copy or share it on other platforms without explicit permission.
Download them in the Files & versions tab.