Flux

Verified against Flux · 2026-07-31

Brief a painterly concept-art illustration on Flux.1 [dev], not a photograph

A non-photographic illustration brief for Flux.1 [dev] that steers the model firmly away from its strong default pull toward photorealism, using explicit medium, brushwork, and technique language to get a genuine painted concept-art look instead of a photo with a style filter over it.

Flux.1 [dev]7 fillable variables

The prompt

Ready to copy — highlighted parts are example details you can swap.

SUBJECT AND SCENE
a lone airship drifting between towering red-rock canyon walls at dusk, tiny lights visible along its hull

MEDIUM
This must render as traditional gouache and ink concept art, painted on textured cold-press paper, not a photograph and not a photorealistic digital render. State this as a positive fact about the image's actual material surface, since Flux's strong default lean toward photographic realism will otherwise reassert itself even after a style is named loosely — the medium has to be described as concretely as a camera and lens would be for a photo brief.

BRUSHWORK AND SURFACE
visible loose, confident gouache brushstrokes with hard ink line reinforcement at key edges. Visible texture matching that brushwork should be evident across the image, especially in the canyon rock faces and the sky, where large flat brush passes should be clearly readable, the way an actual painting shows the physical marks of how it was made.

COLOR PALETTE AND LIGHT
dominant burnt-orange and deep indigo, with the airship's small hull lights as the only warm-white accent. Concept art typically commits to a much more limited, mood-driven palette than a photograph would — describe the two or three dominant colors and how light and shadow are handled within that limited range, rather than full photographic color fidelity.

REFERENCE STYLE
the loose, confident brush economy and high-contrast silhouette work typical of golden-age film concept art, not a tight, airbrushed digital-painting finish. Use this as a description of technique and sensibility — brush size, edge treatment, level of detail versus suggestion — not as an instruction to copy a specific living artist's identifiable signature style; describe the qualities that reference implies rather than naming it as the sole instruction.

COMPOSITION, FOCAL TREATMENT, AND FINISH
the airship rendered with the most linework detail, the canyon walls suggested with broad, looser color shapes, and the frame's corners left with visible unfinished brush edges, the way a concept sketch intentionally doesn't resolve every edge. Concept art typically uses a clear focal hierarchy — one area rendered with real detail, the rest suggested with looser, more gestural marks — rather than the uniform sharp-everywhere detail a photographic brief would ask for, and it is often left with an intentionally unresolved edge or two rather than a fully finished border.

WHAT THE IMAGE SHOULD NOT LOOK LIKE
Because there is no negative-prompt field, state this positively: the entire image should read as the stated medium throughout, including the subject itself — no photographic skin texture, no camera lens characteristics such as bokeh or chromatic aberration or film grain, and no photorealistic material rendering anywhere in the frame.

IF THE RESULT STILL READS AS PHOTOGRAPHIC
If a first attempt still shows photographic lighting or material rendering underneath the named medium, that's a sign the medium and brushwork_description need to be described even more concretely rather than restated more emphatically — naming an additional specific technique detail (a described edge quality, a specific way highlights are rendered as flat color patches rather than a gradient) usually resolves this more reliably than simply repeating "not a photograph" a second time.

CHECK BEFORE FINALIZING
Look specifically at how highlights and reflections are rendered on any smooth or reflective surface in the scene — photographic rendering shows a smooth gradient falloff, while genuine painted media typically render a highlight as a distinct, separately-placed patch of lighter color with a harder edge, and this single detail is often the most reliable tell of whether the medium instruction actually took hold.

OUTPUT
One illustration that a viewer would immediately identify as painted or drawn concept art, not a photo run through a stylistic filter.

Customize

Optional — swap in your own details for the highlighted parts above.

Why this works

Flux's entire training and architecture skew heavily toward photorealistic output — it is, across every published account, one of the strongest photorealism-focused image models available, which is exactly why getting a genuinely painterly, non-photographic result requires actively fighting that default rather than simply naming a style once. A loose instruction like "in the style of a painting" frequently still produces something with photographic lighting and material rendering underneath a thin painterly filter, because the model's strongest prior is photographic, and a single style word is not enough to override it. Describing medium and brushwork with the same concrete specificity a camera-and-lens instruction would get in a photo brief — gouache on cold-press paper, visible loose brushstrokes with ink reinforcement — works for the same underlying reason camera specificity works for photorealism: Flux responds to concrete, describable physical characteristics far more reliably than to abstract style labels. Treating "painting" as a material with actual surface properties, not a filter, gives it something specific to actually render. The composition_and_finish_note distinction — one area rendered in detail, the rest suggested loosely — targets a specific tell that separates real concept art from a photo-to-painting filter effect. Genuine concept art almost always has an intentional focal hierarchy where detail is spent economically, while Flux's photographic default instinct is to render every part of a scene with roughly uniform fidelity, the way a camera lens resolves a whole frame at once. Naming the focal hierarchy explicitly counteracts that uniform-detail default. The instruction to treat a named style_reference as technique-and-sensibility rather than a literal copy instruction matters for two reasons at once: it produces a more genuinely varied, less derivative result than trying to imitate one specific artist's signature marks exactly, and it avoids directly targeting a living artist's identifiable style as the sole instruction, which is why the prompt reframes the reference as a description of qualities — brush size, edge treatment, detail-versus-suggestion — rather than a name to copy wholesale.

Verified against

Flux Flux.1 [dev] · 2026-07-31

Changelog

  • 2026-07-31 Initial publish, verified against Flux.1 [dev] for a gouache-and-ink concept-art brief with an explicit focal hierarchy and finish note.

Need this built into your business?

If a prompt isn't enough — what Scult builds, built and maintained for you — that's Scult's day job.

EXPLORE WHAT SCULT BUILDS
Get a clean, ecommerce-ready product shot from Flux.2 with no negative promptsA structured product-photography brief for Flux.2 that produces a genuinely clean, seamless studio background purely through positive description — because Flux has no negative-prompt field to simply tell it "no clutter, no props, no shadows."Flux.2Flux.1 [pro]2026-07-22Brief Flux.2 for an editorial lifestyle photograph, steered by positive descriptionA camera-and-film-aware natural-language brief for Flux.2 that gets a genuine editorial-magazine look — because Flux has no negative-prompt field, unwanted elements have to be steered out through what you describe, not an exclusion list.Flux.22026-07-24Design a scroll-stopping social graphic on Flux.2 with real reserved caption spaceA platform-aware brief for a social campaign graphic on Flux.2 that reserves genuinely clean space for a text overlay added later in a design tool — achieved by describing the empty area explicitly, since Flux has no exclusion field to keep text or clutter out of it.Flux.22026-08-01Keep a character consistent across multiple scenes with Flux.1 KontextA sequential in-context editing workflow for Flux.1 Kontext that locks a character's face, outfit, and proportions across a series of new scenes by editing forward from one reference image each time, instead of re-describing the character from scratch and getting a visibly different person every generation.Flux.1 Kontext [pro]Flux.1 Kontext [dev]2026-07-26
All Flux prompts

Check your AI visibility

One URL in, a 0–100 score and the exact fixes out.

RUN THE CHECK

Browse all the tools

15 tools across six categories
13 of them never send your data anywhere

Free · No signup · No trial clock

SEE THE DIRECTORY