Krea 2: The Essentials

Covers Krea 2 Medium, Krea 2 Large, Krea 2 Medium Turbo, Krea 2 Turbo, and Krea 2 Turbo Style.

Krea 2 is Krea's second generation image family on Scenario: five models that split one core strength (aesthetic diversity from a natural language prompt) across different priorities, illustration, photorealism, speed, and style transfer, so you pick the tradeoff instead of fighting one model's defaults.

Which Model Should I Use?

Start from the outcome, not the model. Need a specific illustrated or anime look with full control over composition: Krea 2 Medium. Need a photograph-grade result with grain and motion blur: Krea 2 Large. Need either of those fast for quick concepting: the matching Turbo variant. Need to lock in someone else's visual style (a moodboard, a brand look, a piece of concept art) and apply it to a brand new scene: Krea 2 Turbo Style is the only model built for that, since it is the only one where a style reference is mandatory.

How to Use the Model

All five models read a single natural language prompt field, not a list of keywords. Krea 2 responds best to a full descriptive sentence or paragraph that covers the subject, the medium or rendering style, the lighting, and the composition, the way a photographer or art director would brief a shoot. Short, keyword-style prompts under-use the model.

Krea 2 Medium: Illustration and Anime

Medium is the strongest of the five for illustration, anime, and painterly concept art. It exposes aspectRatio (eight presets from square to 2.35:1 cinema) and creativity (raw, low, medium, high) alongside the optional style references.

Krea 2 Medium, aspectRatio 16:9, creativity high: painterly fantasy concept art rewards the higher creativity setting with looser, more atmospheric interpretation · Open on Scenario

Krea 2 Medium, aspectRatio 16:9, creativity high: a glowing underwater fantasy scene, showing the model’s range on non-photographic subject matter · Open on Scenario

Krea 2 Medium with a style reference image (strength 0.8): a children’s book watercolor reference pulled the whole scene into that same soft, painterly palette · Open on Scenario

Krea 2 Large: Photorealism

Large is tuned for camera-true aesthetics: visible grain, motion blur, and natural dynamic range instead of the cleaned-up look most image models default to. Keep creativity low when the brief calls for a specific, believable photograph.

Krea 2 Large, creativity low: motion blur and film grain render as a photographer would actually capture the moment, not as a post-processed effect · Open on Scenario

Krea 2 Large, creativity medium: frozen water droplets and natural light hold up at portrait aspect ratio (9:16) · Open on Scenario

Krea 2 Large, creativity medium: low light interior detail and window glare render with realistic falloff · Open on Scenario

Krea 2 Large with a style reference (cinematic film-grain still, strength 0.7) at aspectRatio 2.35:1: the reference pushed the color grade toward a teal and orange cinematic look · Open on Scenario

Krea 2 Medium Turbo: Fast Illustration

Medium Turbo shares Medium's illustration and anime strengths and its exact parameter set, but generates in a fraction of the time. Reach for it when iterating through several character or poster concepts before committing to Medium for a final pass.

Krea 2 Medium Turbo, aspectRatio 2:3, creativity high: a flat-graphic poster concept generated in seconds · Open on Scenario

Krea 2 Medium Turbo, aspectRatio 16:9, creativity high: dramatic lighting and exaggerated perspective, useful for quick key-art iteration · Open on Scenario

Krea 2 Medium Turbo with a style reference (mid-century travel poster, strength 0.8): the flat three-color palette and poster composition carried straight through · Open on Scenario

Krea 2 Turbo: Fastest General Purpose

Turbo trades the aspectRatio and creativity controls for imageSize presets, an acceleration toggle, and numOutputs up to 4 per run. It has no style reference input at all, it is the pure speed tier, equally comfortable with photoreal and illustrated prompts.

Krea 2 Turbo, imageSize landscape_16_9, acceleration none: a photoreal street scene generated as a single fast pass · Open on Scenario

Krea 2 Turbo, imageSize portrait_4_3, acceleration regular: action and fine fur detail hold up even with acceleration on · Open on Scenario

Krea 2 Turbo, imageSize landscape_16_9: a wide, detail-dense scene rendered in a single fast call · Open on Scenario

Krea 2 Turbo, imageSize landscape_4_3: soft reflections and misty atmosphere from a short, direct prompt · Open on Scenario

Krea 2 Turbo Style: Style Transfer

styleReferenceImages is required here, not optional, one to three images. The model extracts each reference's palette, linework, texture, and composition and applies it to the new scene described in prompt. styleScale (0 to 4, default 1) controls how strongly the references dominate; combining two or three references blends their aesthetics rather than picking one.

Krea 2 Turbo Style, 1 style reference (ukiyo-e woodblock print), styleScale 1: a single strong reference gives a faithful, consistent style match · Open on Scenario

Krea 2 Turbo Style, 2 style references (cyberpunk neon + technical blueprint), styleScale 1.2: two references blend into a combined neon and schematic aesthetic · Open on Scenario

Krea 2 Turbo Style, 3 style references (paper-cutout, felt craft, torn-paper collage), styleScale 0.8: three references at a lower scale produce a softer, blended handmade-craft look · Open on Scenario

Parameters

The five models share a common prompt field but split the rest of the surface by tier. Medium, Large, and Medium Turbo share one parameter set; Turbo and Turbo Style share another, built around resolution presets instead of aspect ratio and creativity.

prompt

Required on every model, up to 5000 characters. Write a full descriptive sentence covering subject, style, lighting, and composition, not a keyword list. See the examples above for the level of detail Krea 2 rewards.

styleReferenceImages

Optional on Medium, Large, and Medium Turbo (up to 10 images, each with its own strength from 0 to 1). Required on Turbo Style (1 to 3 images, no per-image strength, controlled instead by the shared styleScale). Not available on Turbo. The fox-courier, jazz-trio, and lighthouse-keeper examples above show the optional form; the three Turbo Style examples show 1, 2, and 3 references respectively.

styleScale

Turbo Style only. Range 0 to 4, default 1. Higher values push the output closer to the reference images' style at the expense of prompt fidelity. The puppet-show example above used 0.8 to keep three references from overpowering the scene; the hovercar-showroom example used 1.2 for a bolder style match.

aspectRatio

Medium, Large, and Medium Turbo only. Eight presets: 1:1, 4:3, 3:2, 16:9, 2.35:1 (cinema), 4:5, 2:3, 9:16. Default 1:1. The lighthouse-keeper example above uses 2.35:1 for a widescreen cinematic frame.

creativity

Medium, Large, and Medium Turbo only. One of raw, low, medium, high, default medium. Lower values stay close to a literal reading of the prompt, useful on Large when a specific photographic result matters. Higher values let the model take more artistic liberty, useful on Medium and Medium Turbo for looser concept art.

imageSize

Turbo and Turbo Style only. Six presets: square (512x512), square_hd (1024x1024), landscape_4_3, landscape_16_9, portrait_4_3, portrait_16_9. Default square_hd.

numOutputs

Turbo and Turbo Style only. 1 to 4, default 1. Each additional image adds to the cost. The succulent-icon and paper-boat examples in this batch used a batch of 4 to compare variations from a single prompt.

acceleration

Turbo and Turbo Style only. none or regular, default none. The snow-leopard example above used regular without a visible quality tradeoff at that subject and resolution.

enablePromptExpansion

Turbo and Turbo Style only. Boolean, default false. When true, an LLM expands a short prompt into a more detailed one before generation, falling back to the original prompt if expansion fails. Useful for quick, underspecified briefs; skip it when the prompt is already a full detailed paragraph.

seed

All five models. 0 to 2147483647, leave blank for a random seed. On Turbo and Turbo Style, output i in a multi-image batch uses seed + i.

Use Cases

• Game concept art: Medium and Medium Turbo's illustration range covers character sheets, environment key art, and creature design in one prompt-driven pass.

• Marketing and advertising stills: Large's camera-true grain and motion blur produce imagery that reads as shot, not generated, for product and lifestyle campaigns.

• Rapid concepting and pitch decks: Turbo's speed and 4-image batches make it the fastest way to put several visual directions in front of a client in one sitting.

• Brand-consistent content at scale: Turbo Style locks a moodboard or an existing piece of key art in as a style reference, then reapplies it across many new scenes without re-briefing the look each time.

• Editorial and documentary-style imagery: Large at low creativity holds up for photojournalistic and sports subjects where the result has to look captured, not composed.

• Children's and educational publishing: Medium's optional style references can lock a specific storybook illustration style (as in the fox-courier example above) across an entire set of pages.

Tips for Better Results

1. Write a full paragraph, not a keyword list. Every example in this article used a complete descriptive sentence covering subject, medium, lighting, and composition, and that is what Krea 2 was trained to respond to.

2. Match the model to the aesthetic, not the other way around. Forcing a photoreal brief through Medium or an illustrated brief through Large works against each model's tuning.

3. Keep creativity low on Large for anything that has to look like an actual photograph. The marathon-runner and product examples above used low to stay camera-true.

4. Use high creativity on Medium and Medium Turbo for concept art, not final assets. It is a feature for exploring a direction, not for locking one in.

5. On Turbo Style, start at styleScale 1 and adjust from there. Push it above 1 for a bolder style match (the hovercar-showroom example), or below 1 when blending several references so none of them overwhelms the scene (the puppet-show example).

6. Style reference strength is per-image on Medium, Large, and Medium Turbo, but shared on Turbo Style. Do not expect an individual per-image dial on Turbo Style, only the single styleScale.

7. Reach for Turbo first when exploring, then move a winning direction to Medium or Large. Turbo's speed and 4-image batches make it the cheapest way to rule out weak directions before spending on a slower model.

Known Limitations

• No style references on Turbo. If a brief needs a style reference and needs Turbo's speed, Turbo Style is the closer fit, though it makes the reference mandatory rather than optional.

• Turbo Style has no per-image reference strength. Medium, Large, and Medium Turbo let each style reference carry its own strength; Turbo Style only exposes one shared styleScale across all of its references.

• Background text and signage can render illegibly. Like most image models, incidental text in a scene (shop signage, labels) is not guaranteed to be legible; none of these five models are positioned as text-rendering specialists.

• Aspect ratio and resolution controls do not carry across tiers. Medium, Large, and Medium Turbo use named aspect ratios; Turbo and Turbo Style use named resolution presets instead. Swapping between a Turbo-tier and a Medium-tier model in a pipeline means re-specifying framing in a different parameter.