Fast distilled version of Black Forest Labs' FLUX.2 [dev] optimized for speed and cost efficiency.
Settled by community votes across 15 shared challenges, with an AI judge weighing in on each.
FLUX.2 [dev] Flash
#5 of 62 in Text-to-Image
OmniGen v2
#57 of 62 in Text-to-Image
Where the votes landed
FLUX.2 [dev] Flash
100.0%
win rate
Ties
0.0%
OmniGen v2
0.0%
win rate
Challenge by challenge
The strongest take from each model on every shared challenge, with the AI judge's read.
Geometric Composition
Text-to-Image“A glass cube on a wooden table. Inside the cube is a small blue sphere. On top of the cube sits a red book. A green plant is behind the cube, partially visible through the glass. Soft window light from the left.”
AI Judge Analysis
FLUX.2 [dev] Flash
- + Excellent realism with natural textures on the book, wood, and glass.
- + Accurately renders the plant behind the glass with realistic refraction and transparency.
- + Lighting is consistent with the window source, creating a moody and believable atmosphere.
- − The glass cube has slightly thick edges that resemble an aquarium more than a solid glass cube.
OmniGen v2
- + Perfect geometric shapes and clean, vibrant colors.
- + Follows all spatial instructions including the placement of the sphere and book.
- + Minimalist and clean composition.
- − The image looks significantly more 'rendered' and less like a real photograph.
- − The refraction of the plant through the glass is less realistic than Model A.
- − The red book appears to be floating slightly above the glass surface instead of resting on it.
Verdict: FLUX.2 [dev] Flash produces a much more realistic image with convincing lighting, textures, and physics-based refraction through the glass. While OmniGen v2 correctly places all elements, it lacks the photographic depth of FLUX.2 and has a flatter, more digital appearance.
Candid Street Photography
Text-to-Image“A candid street photo of an elderly Japanese man repairing a red bicycle in light rain, reflections on wet pavement, shallow depth of field, 50mm lens, natural skin texture, imperfect framing, motion blur from passing cars, cinematic but realistic, no stylization.”
AI Judge Analysis
FLUX.2 [dev] Flash
- + Excellent skin texture and realistic age details on the subject.
- + Strong adherence to the 'repairing' aspect of the prompt with visible tools.
- + Effective use of motion blur and wet pavement reflections for a cinematic feel.
- − The bicycle frame geometry is slightly warped and inconsistent near the handlebars.
- − The background cars look a bit muddy in their rendering.
OmniGen v2
- + Beautiful, clean reflections on the wet pavement.
- + Composition is balanced with a pleasing color palette.
- − The man is just holding the bike rather than 'repairing' it as requested.
- − The skin looks too smooth and lacks the requested 'natural skin texture' and 'no stylization'.
- − The bike proportions are noticeably off, especially the seat height relative to the frame.
Verdict: FLUX.2 [dev] Flash followed the complex prompt much better, capturing the specific activity of repairing a bike and the grit of a rainy street. OmniGen v2 produced a more stylized, 'clean' image that failed to depict the repairing action and lacked the realistic skin textures requested.
Fantasy Warrior
Text-to-Image“Close portrait of a battle-worn paladin in ornate engraved plate armor, hair braided with small beads, faint scars and dirt on the skin, warm torchlight reflecting off metal, shallow depth of field, bokeh sparks, lifelike eyes, highly detailed texture on leather straps and cloth underlayer.”
AI Judge Analysis
FLUX.2 [dev] Flash
- + Excellent adherence to textural details like the ornate engravings, leather straps, and cloth underlayer.
- + Realistic skin textures with lifelike eyes and grit that matches the 'battle-worn' description.
- + Superior rendering of the 'braided with small beads' hair request.
- − The scars appear more like fresh bloody cuts rather than faint scars.
OmniGen v2
- + Strong cinematic lighting and a pleasing warm color palette.
- + Good composition with a clear focus on the subject and bokeh background.
- − The 'battle-worn' aesthetic is lacking as the character looks perfectly clean aside from a few face dots.
- − The armor and leather textures are much smoother and less detailed compared to Model A.
- − The beads in the hair look like simple white dots rather than integrated decorative beads.
Verdict: FLUX.2 [dev] Flash captures every specific detail of the prompt with high fidelity, particularly the complex textures of the armor, straps, and braided hair. OmniGen v2 produces a beautiful image, but it fails to convincingly portray a 'battle-worn' character and lacks the intricate material depth found in Flux's output.
Modern Clean Menu
Text-to-Image“Modern minimalist restaurant menu design, white background with colorful food photos in grid, sections for appetizers/pizza/mains, bold sans-serif fonts, vibrant accents, clean professional layout for casual dining.”
AI Judge Analysis
FLUX.2 [dev] Flash
- + Excellent photo quality for the food items
- + Follows the vibrant color accents and grid requirement well
- + The layout feels like a genuine, modern professional menu.
- − Text rendering is messy with many typos (e.g., 'CASSUAL', 'PIZZA' vs 'RIZLA')
- − The grid contains nearly identical pizza photos for different categories.
OmniGen v2
- + Structured double-page layout provides a clean sense of organization
- + Better variety of food items (soup, salad, etc.) compared to Model A
- + Modern aesthetic with good use of negative space.
- − Very poor text rendering with many 'hallucinated' characters
- − Large blocks of text are unreadable scribbles
- − Food photos are somewhat smaller and less detailed than Image A.
Verdict: FLUX.2 [dev] Flash produces a much more visually high-end result with superior high-resolution food photography that fits the 'modern minimalist' theme perfectly. While OmniGen v2 has a better variety of food dishes in its grid, its text rendering and clarity are significantly weaker, looking more like an abstract mock-up than a professional product.
Magic Burger Explosion: Fiery Photorealism Challenge
Text-to-Image“Ad for 'Magic Burger'. Dynamic, exploded burger with all components (bun, patty, cheese, lettuce, tomato, sauce) suspended in mid-air. Emphasize photorealistic detail and a sense of motion. Dark, fiery background with glowing embers. Integrate text: 'MAGIC BURGER' as a prominent title, 'LIMITED TIME ONLY' as a secondary message, and '€6.99' in a starburst, all rendered with a fiery, glowing effect.”
AI Judge Analysis
FLUX.2 [dev] Flash
- + Excellent adherence to the 'exploded' and 'suspended in mid-air' requirement.
- + Highly realistic food photography texture and lighting.
- + Perfect text rendering with the specified fiery, glowing effect.
- − The 'MAGIC BURGER' text is slightly less central than a traditional logo might be.
OmniGen v2
- + Strong, high-contrast graphic design suitable for a cartoonish advertisement.
- + Clear and legible main logo text.
- − Failed to create an exploded burger; the components are stacked normally.
- − Text rendering is slightly clipped on 'LIMITED TIME ONLY' and missing the Euro symbol.
- − Artificial, 'plastic' look rather than the requested photorealistic detail.
Verdict: FLUX.2 [dev] Flash perfectly followed the prompt's structural requirements, creating a dynamic exploded burger with a dark, sophisticated, and realistic aesthetic. OmniGen v2 failed the primary 'exploded' instruction, producing a standard stacked burger with less polished text and a graphics-heavy, non-photorealistic style.
Chalkboard Menu
Text-to-Image“Handwritten-style chalkboard menu in a cozy café, all text rendered in the exact same realistic chalk handwriting style with natural variations in letter size, slight slant, and chalk texture — no printed or digital fonts anywhere on the board. Title at the top in elegant cursive chalk handwriting: ‘TODAY’S SPECIALS – APRIL 30, 2026’. Below it, three menu items also in the same handwritten chalk style: ‘Truffle Mushroom Risotto – $24’, ‘Grilled Octopus with Lemon & Herbs – $28’, ‘Brown Butter Chocolate Chip Cookies – $9’. At the very bottom, smaller text in the identical handwritten chalk style (slightly smaller but still clearly legible with the same handwriting characteristics): ‘All items made fresh daily • Ask about our gluten-free options’. Warm ambient café lighting, visible chalk dust and smudges, realistic handwriting imperfections, no clean printed text anywhere.”
AI Judge Analysis
FLUX.2 [dev] Flash
- + Perfect text rendering for all requested menu items with no spelling errors.
- + Authentic chalk texture and natural handwriting variations.
- + Excellent atmospheric lighting and background depth for a cozy café feel.
- − The 'Today's Specials' title uses a more printed-block style rather than the requested 'elegant cursive'.
OmniGen v2
- + The font for the date and small text captures a casual chalk-marker aesthetic well.
- + Good use of frame-style composition.
- − Numerous spelling errors and garbled text including 'SPECALS' and 'Trufa Munst'.
- − Fails to render the request for a 'cozy café' background, instead placing the board on a flat surface.
- − Messy and unreadable handwriting in the middle section of the menu.
Verdict: FLUX.2 [dev] Flash significantly outperforms OmniGen v2 by providing perfectly spelled, high-resolution text and a realistic café environment. While OmniGen v2 struggles with basic spelling and letter formation, FLUX.2 [dev] Flash delivers a professional-grade chalkboard menu that closely adheres to the complex text requirements of the prompt.
The Reversed Rodeo
Text-to-Image“Horse riding astronaut in space — horse on top, not vice versa. Surreal, highly detailed, cinematic.”
AI Judge Analysis
FLUX.2 [dev] Flash
- + Excellent cinematic lighting and atmospheric depth
- + High level of detail in the astronaut's suit and the horse's texture
- + Strong composition with a sense of motion
- − Failed the negative constraint; the astronaut is riding the horse instead of vice versa
OmniGen v2
- + Clean, simple illustration style
- + Correctly identifies the core subjects
- − Failed the negative constraint; the astronaut is riding the horse instead of vice versa
- − Lacks the requested cinematic and surreal qualities
- − Anatomical issues where the horse's back legs blend into the background
Verdict: Both FLUX.2 [dev] Flash and OmniGen v2 failed the specific negative constraint to have the horse on top of the astronaut. However, FLUX.2 [dev] Flash produced a much higher quality image that accurately captured the cinematic and highly detailed aesthetic requested, whereas OmniGen v2 delivered a flatter, less coherent graphic.
The Capybara Taxi Driver
Text-to-Image“Photorealistic scene inside a yellow New York taxi at night. A capybara is driving, wearing a yellow taxi driver cap and a dark jacket. It has a calm, professional expression and both front paws on the steering wheel. In the back seat sits a human businesswoman in a coat, looking at her phone with a completely normal, bored expression (as if this is just another normal ride). Through the windows you can see the streets of Manhattan at night with blurred lights. Realistic taxi interior, photorealistic, detailed fur and fabric, 35mm lens, night lighting with reflections, shallow depth of field.”
AI Judge Analysis
FLUX.2 [dev] Flash
- + Excellent adherence to the 'both front paws on the steering wheel' instruction with realistic capybara anatomy.
- + Highly photorealistic textures on the capybara's fur and the passenger's clothing.
- + Captures the bored expression of the passenger perfectly while placing her correctly in the back seat.
- − The capybara's claws are slightly exaggerated/hook-like on the wheel.
- − Composition is a bit crowded from the front-facing angle.
OmniGen v2
- + Clean, vibrant colors and sharp focus on the subjects.
- + The businessman's suit looks high-quality and professional.
- − Failed the main prompt instruction by giving the capybara human hands on the steering wheel.
- − The passenger is sitting in the front passenger seat instead of the requested back seat.
- − The image has a more digital, smoothed-over 'AI look' rather than the requested photorealistic style.
Verdict: FLUX.2 [dev] Flash followed the complex prompt instructions perfectly, correctly placing the woman in the back seat and giving the capybara realistic paws for driving. OmniGen v2 failed significantly by giving the animal human hands and misplacing the passenger in the front seat. FLUX.2 also achieved a much higher level of photorealism.
The Halloween Invitation
Text-to-Image“Vintage gothic Halloween party invitation. Dark parchment poster, spooky border with webs and thorns, central glowing jack-o-lantern, bats, twisted trees, moody night sky. Add elegant gothic title text saying "Halloween Party Invitation", a small scroll banner saying "You are invited to a night of frights", and event details at the bottom: Date: 30.10.2026 Time: 7pm Location: The Arches, NYC Spooky but polished, cinematic lighting, square format.”
AI Judge Analysis
FLUX.2 [dev] Flash
- + Excellent text rendering with no spelling errors.
- + Sophisticated cinematic lighting and moody gothic atmosphere.
- + Strong adherence to all prompt elements including border details like webs and thorns.
- − Includes a small line of garbled text ('Hurk: O369') in the event details.
OmniGen v2
- + Strong 'parchment' aesthetic for the poster backing.
- + High-contrast color palette with a glowing moon effect.
- − Significant text mangling and spelling errors in the scroll banner and details.
- − Visual style leans towards clip-art rather than 'polished cinematic' lighting.
- − Layout is cluttered with overlapping and repetitive text.
Verdict: FLUX.2 [dev] Flash is the clear winner as it successfully rendered almost all text perfectly and captured the requested 'cinematic' and 'gothic' atmosphere. OmniGen v2 struggled significantly with text legibility and produced a much coarser, less professional illustration style.
Isometric Miniature Diorama Scenes
Text-to-Image“Create a clear, 45° top-down isometric miniature 3D cartoon scene of Japan's signature dish: sushi, with soft refined textures, realistic PBR materials, gentle lighting, on a small raised diorama base with minimal garnish and plate. Solid light blue background. At top-center: 'JAPAN' in large bold text, 'SUSHI' below it, small flag icon. Perfectly centered, ultra-clean, high-clarity, square format.”
AI Judge Analysis
FLUX.2 [dev] Flash
- + Excellent text rendering and alignment with the requested layout
- + High-quality PBR textures on the wood and fish
- + Accurate Japanese flag icon
- − The sushi composition is a bit unusual, mixing nigiri toppings on maki-style rolls
OmniGen v2
- + Strong 3D cartoon aesthetic with vibrant colors
- + Good isometric perspective and diorama base styling
- − Incorrect flag icon (red/yellow stripes instead of Japanese flag)
- − Slightly muddy text rendering with shadows overlapping
- − Nigiri fish has an anatomical tail-like artifact reaching onto the plate
Verdict: FLUX.2 [dev] Flash followed the prompt instructions much more accurately, particularly regarding the text placement and the specific flag icon. While OmniGen v2 captured the 3D cartoon style well, it failed at representing the Japanese flag and had structural issues with the sushi model.
Adorable Baby Animals in Sunny Meadow
Text-to-Image“Hyper-photorealistic scene of fluffy baby animals—a golden retriever puppy, tabby kitten, baby bunny, and red fox kit—with big expressive eyes and ultra-detailed soft fur, playfully chasing butterflies and tumbling together in a lush wildflower meadow, warm golden sunrise light with god rays and dew sparkles, joyful wholesome vibe, 8K masterpiece.”
AI Judge Analysis
FLUX.2 [dev] Flash
- + Excellent adherence to the 'hyper-photorealistic' instruction with realistic fur textures.
- + Successfully included all requested animal types including the fox, kitten, puppy, and bunnies.
- + Beautiful lighting effects including god rays and detailed dew drops on grass.
- − Generated extra animals beyond the singular list (two bunnies and two foxes).
OmniGen v2
- + Bright, vibrant colors that convey a 'joyful vibe'.
- + Clear, centered composition.
- − Failed the 'photorealistic' requirement, delivering a 3D cartoon/animation style instead.
- − Missing the baby bunny mentioned in the prompt.
- − Anatomy is highly stylized and lacks realistic detail in fur and features.
Verdict: FLUX.2 [dev] Flash followed the stylistic requirements perfectly, producing a stunningly realistic image with intricate fur and lighting details. OmniGen v2 failed to achieve photorealism, providing a cartoonish output that also ignored the request for a baby bunny. FLUX.2 is the clear winner for its superior texture, atmosphere, and prompt adherence.
Studio Ghibli Anime Style
Editing“Transform this photo into a Studio Ghibli–inspired illustration. Use soft pastel colors, hand-painted textures, gentle lighting, dreamy backgrounds, and a warm, nostalgic mood”
AI Judge Analysis
FLUX.2 [dev] Flash
- + Excellent capture of the Ghibli 'watercolour' background aesthetic with soft pastel tones.
- + Perfectly preserves the composition and specific outfits (plaid pattern, blue tank top) of the source image.
- + Accurately replicates the character expressions, including the jealous look of the woman on the right.
- − The line art is a bit faint, almost losing some of the character definition in the bright light.
OmniGen v2
- + Successfully translates the scene into a clean anime art style with bold line work.
- + Retains the layout and environmental context (street, bus) from the original photo.
- + Character features are distinct and consistent with a generic modern anime look.
- − Fails to capture the specific 'Ghibli' aesthetic, looking more like a standard vector-style anime.
- − Completely changes the expressions of the man and the woman on the right, losing the 'jealous' and 'distracted' emotional context of the meme.
Verdict: FLUX.2 [dev] Flash is the clear winner as it masterfully blends the iconic Ghibli hand-painted texture and soft lighting with the specific details of the source image, including the crucial facial expressions. OmniGen v2 creates a generic anime version that loses the emotional nuance of the original meme and misses the specific artistic requirements of the 'Ghibli' prompt.
Golden Hour Stroll
Image Editing“Add dynamic motion to this photo: make hair blow in the wind, add leaves flying, energetic and lively feel.”
AI Judge Analysis
FLUX.2 [dev] Flash
- + Excellent preservation of the original person, dog, and background details.
- + Highly effective interpretation of wind on the hair while maintaining facial structure.
- + The leaves feel integrated into the 3D space with varying sizes and blurs.
- − Some leaves appear to be overlapping the woman's torso in a slightly unnatural way.
OmniGen v2
- + Successfully added blowing hair and flying leaves.
- + Maintains a clean, high-saturation aesthetic.
- − Significantly altered the facial features and hair color of the woman compared to the source.
- − The flying leaves look like simple 2D overlays rather than part of the environment.
- − The dog's face has been noticeably changed/smoothed compared to the original.
Verdict: FLUX.2 [dev] Flash is the clear winner as it successfully follows all edit instructions while perfectly preserving the identity of the person and the dog. OmniGen v2 failed the primary task of source preservation, essentially generating a new character and dog that resemble the originals rather than editing the actual photo.
Vintage Cafe Logo
Text-to-Image“Vintage minimalist restaurant logo for "Caffè Florian", retro cloche dome with steam and "Est. 1720" banner, classic typography, warm brown and cream tones, subtle texture on light background, vector emblem style.”
AI Judge Analysis
FLUX.2 [dev] Flash
- + Excellent typography including the requested accent mark.
- + Superior subtle texture that perfectly matches the vintage prompt.
- + Professional vector lighting and shading on the cloche dome.
- − The steam is a bit stylized and thick compared to the rest of the logo.
OmniGen v2
- + Clean, high-contrast minimalist design.
- + Accurate 'Est. 1720' text rendering.
- − Spelling error in the main brand name ('CAFFFLORIN').
- − Flat composition lacks the requested vintage depth and texture.
- − Steam is overly simplified.
Verdict: FLUX.2 [dev] Flash significantly outperformed OmniGen v2 by providing a professional-grade logo with accurate spelling, specialized typography, and a nuanced vintage texture. OmniGen v2 failed the primary text requirement with a spelling error and missed the artistic subtlety requested in the prompt.
Apollo 11: Journey to Tranquility
Text-to-Image“Create a clean, modern vector infographic poster about the Apollo 11 mission. NASA-inspired palette (navy, white, muted red, light gray). Flat-vector style, crisp lines, consistent iconography, subtle gradients only. Steps (stop at landing): 1. Launch (Saturn Vicon) 2. Earth Orbit (Earth + orbit ring icon) 3. Translunar (trajectory arc icon) 4. Lunar Orbit (Moon + orbit ring icon) 5. Descent (lunar module descending icon) 6. Landing (lunar module on the surface icon) Small supporting elements (minimal text): • Crew strip: three silhouette icons with only last names: Armstrong, Aldrin, Collins. • Landing site marker: Moon pin labeled "Tranquility" only. Layout constraints: generous margins, large readable labels, clean background with subtle stars. Vector-only, print-poster look, high resolution.”
AI Judge Analysis
FLUX.2 [dev] Flash
- + Strong text legibility for the main labels and astronaut names.
- + High-quality, detailed icons that accurately reflect the mission hardware.
- + Excellent adherence to the requested NASA-inspired color palette and vector style.
- − The layout is a bit cluttered with redundant labels for 'Lunar Orbit' and 'Descent'.
- − The Saturn V rocket depicts three stages/boosters side-by-side rather than a single tower.
OmniGen v2
- + Successfully captures a clean grid-based infographic layout.
- + Adheres well to the flat-vector design aesthetic requested.
- − Frequent spelling errors including the main title 'APOLO 17' (requested Apollo 11) and agency name 'NSA'.
- − Icons are abstract and do not represent the specific requested mission steps effectively.
- − Text is largely gibberish or heavily misspelled throughout the poster.
Verdict: FLUX.2 [dev] Flash is the clear winner as it produced a highly legible, accurate, and detailed infographic that followed almost all prompt instructions perfectly. In contrast, OmniGen v2 failed on text accuracy, mission facts (labeling it Apollo 17), and the specific sequence of requested icons.
Explore each model
Unified multimodal model for text-to-image generation, instruction-guided image editing, personalized generation, and virtual try-on