Style Preset (Raw Style)
A style preset is a pre-packaged aesthetic toggle that forces an image generator toward a specific visual category, such as commercial illustration, photography, anime, or 3D art. Instead of relying entirely on text descriptions, picking a preset alters how the model renders lighting, texture, and composition. In tools like Midjourney, a raw preset strips out the default cinematic house style to produce more literal, grounded visuals.
Also known as: style_type, style_preset, --style raw, --raw
Side-by-side test results




Model: Ideogram 3.0 (przez API fal.ai, tryb TURBO). Settings: seed 1411, prompt expansion disabled, only style changes. Date: 14.11.2026.
Prompt: illustration of a small mountain hut in winter, snow, pine trees, smoke from the chimney
In my test run, AUTO and DESIGN produced nearly identical flat illustrations, as the model assumed the word 'illustration' automatically meant graphic design. GENERAL pushed out a broader landscape with mountains in the background, while REALISTIC muted the palette and shifted toward a foggy photograph despite the word 'illustration' remaining in the prompt.
Syntax across generators
| Generator | Parameter | Values | Default |
|---|---|---|---|
| Midjourney* | --style raw / --rawRaw usuwa domyślną stylizację, lepsza zgodność z promptem | - | - |
| FLUX | raw (Ultra) - see stylization | true/false | false |
| Ideogram | style_type / style_presetstyle_preset does not combine with style_codes or style_reference_images. Ideogram 4.5: no style_type | style_type (3.0): auto, general, realistic, design, custom, fiction, stylized. style_preset (3.0): 62 values: 80s_illustration, 90s_nostalgia, abstract_organic, analog_nostalgia, art_brut, art_deco, art_poster, aura, avant_garde, bauhaus, blueprint, blurry_motion, bright_art, c4d_cartoon, childrens_book, collage, coloring_book_i, coloring_book_ii, cubism, dark_aura, doodle, double_exposure, dramatic_cinema, editorial, emotional_minimal, ethereal_party, expired_film, flat_art, flat_vector, forest_reverie, geo_minimalist, glass_prism, golden_hour, graffiti_i, graffiti_ii, halftone_print, high_contrast, hippie_era, iconic, japandi_fusion, jazzy, long_exposure, magazine_editorial, minimal_illustration, mixed_media, monochrome, nightlife, oil_painting, old_cartoons, paint_gesture, pop_art, retro_etching, riviera_pop, spotlight_80s, stylized_red, surreal_collage, travel_poster, vintage_geo, vintage_poster, watercolor, weird, woodblock_print | style_type: general (3.0 Character: auto) |
| Stability AI (SD 3.5) | style_presetSD 3.5, Ultra, inpaint, outpaint, creative upscale, control | enhance, anime, photographic, digital-art, comic-book, fantasy-art, line-art, analog-film, neon-punk, isometric, low-poly, origami, modeling-compound, cinematic, 3d-model, pixel-art, tile-texture (17) | none |
| GPT Image | none in retrieved guide | - | - |
* Midjourney: docs.midjourney.com blocks automated scrapers, so values are sourced from search engine indexes. Please verify settings on the official Midjourney site before production use.
- Sources, verified on 14.11.2026:
- Midjourney: docs.midjourney.com/hc/en-us/articles/48495453462797-Edit-M...
- FLUX: docs.bfl.ml/api-reference/models/generate-an-image-with-flu...
- Ideogram: developer.ideogram.ai/openapi/ideogram-api-v2.json
- Stability AI: docs.stability.ai/api-reference/generate/stable-diffusion-3...
- GPT Image: developers.openai.com/api/docs/guides/image-generation
When to use it
- When an engine stubbornly generates a glossy photograph but you need vector clip art for a community flyer or a local bakery logo
- Inside Stability AI APIs, where you can select from around 17 ready-made presets like photographic, anime, or pixel-art
- Inside Midjourney, when using --style raw to reduce automatic embellishments and follow complex text prompts more faithfully
Common mistake
The most common misstep is introducing a direct conflict between your chosen preset and your written text, like pairing REALISTIC with 'cartoon illustration'. The engine struggles to balance two opposing directives, yielding an awkward, muddy compromise. Another frequent mistake is digging through menus for style_type in Ideogram 4.5, where the parameter is not supported according to official documentation.
Ready-to-use example
Ideogram 3.0: "style_type": "DESIGN" or "style_preset" (62 values) Stability: "style_preset": "photographic" Midjourney: cabin in the mountains --style raw
Frequently asked questions
What does --style raw do in Midjourney?
It dials back the engine's default aesthetic processing. Your results look less like curated cinematic art and adhere more strictly to your written keywords.
Can I use style presets alongside descriptive art style keywords?
Yes, provided they do not contradict each other. For example, pairing a graphic preset with 'screenprint poster' works smoothly, whereas pairing it with 'hyperrealistic photo' causes visual conflicts.
Does using a raw preset make image quality worse?
Not at all. It simply disables automatic beautification, giving you neutral lighting, natural skin textures, and greater manual control over the final render.
