CivArchive
    Raz's 80s Fantasy Movie Style [ZIB+ZIT] - 80s-FantasyMovie-Mk.1
    NSFW
    Preview 141055022
    Preview 141090083
    Preview 141090087
    Preview 141055550
    Preview 141055753
    Preview 141061334
    Preview 141061338
    Preview 141055572
    Preview 141090084
    Preview 141055577
    Preview 141055581
    Preview 141061323
    Preview 141090092
    Preview 141090088
    Preview 141061337
    Preview 141055643
    Preview 141061336
    Preview 141055676
    Preview 141055760
    Preview 141055830

    If you feel like showing some love, awesome.

    But the best kind of support?

    Use the model. 🛠️

    Post your results. 🖼️

    Let the art speak. 🎯


    ⚔️ Raz’s 80s Fantasy Movie Style

    Raz’s 80s Fantasy Movie Style is my first LoRA created for Z Image Base.

    The goal was to capture the feeling of an old fantasy movie discovered on a worn VHS tape: painted sets, practical effects, theatrical lighting, handmade costumes, misty forests, torch-lit castles, strange creatures, dramatic heroes, and that slightly imperfect analog charm.

    Think classic sword-and-sorcery adventures, dark fairy tales, enchanted forests, mysterious ruins, practical creature effects, and fantasy worlds that feel physically built rather than digitally perfect.

    This is not meant to recreate one specific movie. It is my interpretation of the broader 1980s fantasy-film atmosphere: dramatic, colorful, mysterious, handcrafted, and occasionally a little strange.

    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

    This LoRA was created to be used with my:
    Basic T2I — Z Image Base + Z Image Turbo Workflow


    The workflow uses Z Image Base (ZIB) for the main generation, providing better seed variation, composition, and flexibility while avoiding some of the rigidity and style bias of Turbo.

    Z Image Turbo (ZIT) is then used as a low-denoise second pass to enhance fine details, photoreal textures, and overall image quality.

    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

    What It Does

    The LoRA can add:

    • Old-school fantasy movie atmosphere

    • Practical set and creature-effect aesthetics

    • Theatrical costumes and makeup

    • Painted backdrops and handcrafted environments

    • Warm torchlight, colored rim lighting, fog, and haze

    • Analog film texture and subtle grain

    • Dramatic sword-and-sorcery compositions

    • Dark fairy-tale and enchanted-forest moods

    • A less polished, more physical and cinematic fantasy world

    It works well for heroes, villains, witches, knights, barbarians, creatures, castles, caves, ancient temples, enchanted forests, and other classic fantasy subjects.

    You can simply describe the scene you want and let the LoRA establish the visual language.

    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

    🎬 The Feel I Wanted

    I wanted images that feel like frames from a forgotten fantasy movie rented from a video store decades ago.

    Not modern, ultra-clean fantasy.

    Something more theatrical and handmade, where the costumes have texture, the locations feel like sets, the creatures look physically present, and the lighting carries as much personality as the characters.

    The LoRA can move between colorful adventure, dark fantasy, sword and sorcery, and strange fairy-tale imagery depending on the prompt.

    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

    🧙 Prompt Hints

    The style should appear naturally, but you can reinforce specific directions with terms such as:

    1980s fantasy movie, sword and sorcery, dark fairy tale, practical effects, theatrical costume, painted backdrop, analog film, film grain, misty forest, torchlight, ancient castle, fantasy creature, dramatic lighting

    For a cleaner result, keep the prompt simple and focus on the subject, environment, and mood.

    For a stronger retro-film look, describe the lighting, practical effects, costumes, set design, fog, and analog texture more directly.

    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

    🔧 My Personal Settings

    Resolution: 1088 × 1920
    ModelSamplingAuraFlow Shift: 7

    Z Image Base — Main Generation

    Steps: 50
    CFG: 4.00
    Sampler/Scheduler: Euler + Beta
    Denoise: 1.00
    LoRA Strength: 1.00

    Z Image Turbo — Detail Pass

    Steps: 4 or 8
    CFG: 1.00
    Sampler/Scheduler: Euler + Beta
    Denoise: 0.15–0.20
    LoRA Strength: 1.00

    The negative prompt is zeroed out for the ZIT pass.

    4 steps:
    A subtle detail enhancement that stays closer to the original ZIB result.

    8 steps:
    Stronger detail, texture, and realism enhancement. Depending on the image, this can sometimes push the result further than intended.

    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

    🎚️ Strength Notes

    Z Image Base

    1.00 is my recommended starting strength and the value used for my showcase images.

    Z Image Turbo

    The LoRA also works directly with Z Image Turbo, although Turbo may require a strength higher than 1.00 for the style to become equally visible.

    Start at 1.00 and increase gradually if the influence feels too subtle.

    Dual ZIB + ZIT Workflow

    When using my dual-model workflow, I apply the LoRA to both model paths:

    • ZIB LoRA Strength: 1.00

    • ZIT LoRA Strength: 1.00

    Because ZIT is only performing a low-denoise detail pass, both can remain at full strength without completely rebuilding the original composition.

    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

    ⚠️ Notes

    • This is my first Z Image Base LoRA, so Mk.1 is also the beginning of my exploration of training directly for ZIB.

    • The intensity of the style can change depending on the subject and how strongly the prompt leans toward modern photography or modern digital fantasy.

    • Z Image Turbo may show a lighter style influence at 1.00 and can require a higher LoRA strength.

    • Heavy ZIT denoise or an aggressive 8-step second pass can reduce some of the softer analog character created during the Base generation.

    • Simple, descriptive prompts often leave more room for the LoRA’s atmosphere to emerge.

    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

    🧠 In Short

    Handmade fantasy, theatrical lighting, practical magic, and the atmosphere of a forgotten 1980s movie.

    A little dramatic.
    A little strange.
    And hopefully full of adventure.

    Description

    Mk.1 - Original Release

    My first LoRA trained for Z Image Base, built around the atmosphere of classic 1980s fantasy cinema.

    Recommended strength:

    • Z Image Base: 1.00

    • Z Image Turbo: 1.00 or higher if needed

    • Dual ZIB + ZIT workflow: 1.00 on both model paths

      A trigger word is not required for the LoRA to work, but there is a slight difference between results generated with and without it.

      Try both approaches... the difference is subtle, and which one works best will depend on the image and direction you want to explore.

    Designed and tested primarily with my Basic T2I ZIB + ZIT workflow.

    LORA
    ZImageBase

    Details

    Downloads
    87
    Platform
    CivitAI
    Platform Status
    Available
    Created
    8/28/2026
    Updated
    9/7/2026
    Deleted
    -
    Trigger Words:
    Raz's-80sFantasyMovie-Mk1

    Files

    ZIB+ZIT-Raz's-80sFantasyMovie-Mk.1.safetensors