Black Forest Labs' aesthetically-tuned 12-billion parameter flow transformer optimized for high-quality images with incredible aesthetics, suitable for personal and commercial use
Settled by community votes across 13 shared challenges, with an AI judge weighing in on each.
FLUX.1 Krea [dev]
#48 of 62 in Text-to-Image
FLUX.2 [dev]
#19 of 62 in Text-to-Image
Where the votes landed
FLUX.1 Krea [dev]
0%
win rate
Ties
0%
FLUX.2 [dev]
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.1 Krea [dev]
- + Excellent photorealism with convincing material textures
- + Beautiful depth of field and soft lighting
- + Sophisticated reflections on the glass and blue sphere
- − The glass cube has a thick base that looks more like a tray than a pure cube
- − The plant behind the cube is very dark and lacks clarity
FLUX.2 [dev]
- + Perfect adherence to the spatial requirements of the prompt
- + Clear visibility of the green plant through the glass
- + Bright, realistic window light from the left
- − The blue sphere looks slightly more plastic/less detailed than in Image A
Verdict: Both models followed the complex spatial prompt perfectly. FLUX.1 Krea [dev] produces a more artistic, moody photograph with superior textures, while FLUX.2 [dev] provides a clearer, brighter image that better showcases the specific detail of the plant being seen through the glass.
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.1 Krea [dev]
- + Strong cinematic lighting and atmosphere.
- + Accurate reflection of headlights on the wet pavement.
- + Clear visual of the red bicycle frame.
- − The man is just holding the bike rather than 'repairing' it.
- − Skin texture feels slightly smoothed or 'AI-like' compared to the other model.
- − The background cars look static despite the prompt asking for motion blur.
FLUX.2 [dev]
- + Excellent adherence to the 'repairing' action with focused hands.
- + Realistic skin texture and lifelike hair details.
- + Effective motion blur on the passing vehicle in the background.
- − The anatomy of the hands and the number of tools/dials on the handlebars is a bit cluttered and nonsensical.
- − The composition is a bit tight on the subject's face.
Verdict: FLUX.2 [dev] significantly outperformed FLUX.1 Krea [dev] in following the specific requirements of the prompt, particularly the 'repairing' action and the 'motion blur' of passing cars. While FLUX.1 Krea [dev] created a more balanced cinematic composition, FLUX.2 [dev] delivered superior realism in skin texture and atmosphere.
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.1 Krea [dev]
- + Excellent high-contrast cinematic lighting.
- + Very intricate and clean engraving on the plate armor.
- + Superior skin texture and realistic facial expression.
- − The beads in the hair are less prominent and colorful compared to Model B.
- − The 'blood' or dirt on the face looks slightly repetitive.
FLUX.2 [dev]
- + Excellent adherence to the request for colorful beads in the braids.
- + Included the actual torch in the frame which justifies the lighting source better.
- + Great detail on the leather straps and buckles as requested.
- − The facial scars look a bit like physical gashes rather than 'faint scars'.
- − The engraving on the armor is less sharp and looks slightly more 'stamped' than etched.
Verdict: Both models followed the prompt exceptionally well, but FLUX.1 Krea [dev] produces a more cinematic and realistic texture on the skin and metal. FLUX.2 [dev] followed the specific detail regarding beads and leather straps more literally, though the overall image feels slightly more digital and less like a high-end photograph compared to FLUX.1 Krea [dev].
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.1 Krea [dev]
- + Excellent professional layout with a logical single-column text/grid-photo split.
- + Strong consistency in font weights and hierarchy.
- + Very clean minimalist aesthetic that feels like a real restaurant menu.
- − Repeats the 'Appetizers' section heading twice.
- − Small artifacts present in some text lines.
FLUX.2 [dev]
- + More colorful and vibrant use of accents.
- + Better inclusion of specific categories requested like 'Pizza'.
- + Larger, more appetite-appealing food photography.
- − Composition feels a bit cluttered and unbalanced between columns.
- − Garbled text rendering in several titles (e.g., 'PIZZAU').
- − Mixed styling for prices and descriptions looks less professional.
Verdict: FLUX.1 Krea [dev] produces a more realistic and professionally balanced layout that strictly adheres to the 'minimalist' and 'clean' aspects of the prompt. While FLUX.2 [dev] offers more vibrant colors and better food photography, its chaotic typography and spelling errors distract from the overall design.
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.1 Krea [dev]
- + High level of photorealistic detail on the burger components and textures.
- + Correctly features the 'LIMITED TIME ONLY' and 'MAGIC BURGER' text with clear legibility.
- + Dynamic composition with realistic embers and fire beneath the burger.
- − Failed to put the price inside a starburst; placed it at the bottom in a plain font instead.
- − The 'fiery, glowing effect' on the text is very subtle compared to the request.
- − Includes some hallucinated gibberish text at the very bottom.
FLUX.2 [dev]
- + Perfectly followed the text styling instructions, including the fiery glow and the price within a starburst.
- + Strong sense of motion with liquid sauce droplets suspended in mid-air.
- + Cohesive color palette where the lighting from the fire interacts well with the burger and text.
- − The burger bun and lettuce appear slightly more 'illustrated' and less photorealistic than in the competing image.
- − The spacing between the top bun and the rest of the burger is a bit wide, making it look slightly disconnected.
Verdict: FLUX.2 [dev] is the clear winner as it followed every specific detail of the prompt, particularly the complex text requirements like the glowing fiery effect and the price starburst. While FLUX.1 Krea [dev] produced a slightly more realistic burger patty, it failed to integrate the price according to the instructions and included messy artifact text at the bottom.
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.1 Krea [dev]
- + Excellent layout with a clean wood frame
- + Follows the general prompt instructions for dates and items
- − Numerous spelling errors including 'Gruffle' and 'Clvrucuations'
- − Text looks like a digital font rather than realistic chalk with varied pressure
- − Failed to complete the 'Brown Butter' item correctly
FLUX.2 [dev]
- + Superb chalk texture with realistic powdery effects and smudges
- + Perfect spelling for all requested menu items
- + Successfully interprets the handwriting style with natural variations in letter size and slant
- − Framing is slightly tight at the top of the board
- − Small smudge artifact near 'with' on the second item
Verdict: FLUX.2 [dev] is the clear winner as it perfectly executed the 'chalkboard' aesthetic with authentic textures and flawless spelling of the specific menu items. FLUX.1 Krea [dev] failed on both technical handwriting realism and basic spelling, resulting in a more digital-looking image with nonsensical text.
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.1 Krea [dev]
- + Excellent handling of the horse's anatomy integrated with space gear.
- + Strong cinematic lighting with realistic reflections on Earth and the helmet.
- + Good contrast between the white horse and the dark space background.
- − Failed the specific spatial instruction for the horse to be 'on top' of the astronaut.
- − Anatomy issue where the horse's front legs appear to be emerging from a mechanical box.
FLUX.2 [dev]
- + Beautiful background with nebulae and a moon for a more surreal feel.
- + Consistent textures on the horse's fur and the astronaut's suit.
- + Dynamic sense of motion with the horse's mane and tail.
- − Failed the specific spatial instruction for the horse to be 'on top' of the astronaut.
- − Minor anatomical glitch with the horse's back leg merging into the atmosphere/dust.
Verdict: Both models failed the negative constraint/spatial instruction to have the horse on top of the astronaut, instead defaulting to the more common image of an astronaut riding a horse. FLUX.1 Krea [dev] is preferred because its composition feels more cinematic and clear, whereas FLUX.2 [dev] has a slightly cluttered background and less interesting lighting on the main subject.
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.1 Krea [dev]
- + Excellent atmospheric lighting and depth of field.
- + The capybara's fur texture is very detailed and realistic.
- + Characters are well-positioned within the vehicle frame.
- − Includes an extra person in the front passenger seat not mentioned in the prompt.
- − The 'TAXI' hat looks like a cheap plastic prop rather than a driver's cap.
FLUX.2 [dev]
- + Perfectly depicts the capybara's front paws on the steering wheel.
- + The businesswoman's bored expression perfectly matches the prompt's tone.
- + The lighting inside the car feels grounded and photorealistic.
- − The capybara's face is slightly blurry compared to its body texture.
- − The reflection in the windshield is a bit busy and distracting.
Verdict: FLUX.2 [dev] is the winner because it accurately followed the composition of the prompt, specifically involving only two characters and correctly placing the capybara's paws on the wheel. While FLUX.1 Krea [dev] has beautiful lighting, it hallucinated a second human passenger and failed the specific paw placement requested.
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.1 Krea [dev]
- + Excellent atmospheric lighting on the central jack-o-lantern
- + Highly intricate spider web and ornate corner borders
- + Elegant use of multiple ribbon banners for different text sections
- − Major spelling errors including 'Pasty Halloween' and 'You are are invited to a nights Tright'
- − Inconsistent date and time rendering compared to the prompt
FLUX.2 [dev]
- + Flawless text rendering exactly matching the prompt's content
- + Strong composition with a clear thorn-wrapped border
- + Excellent legibility for the event details at the bottom
- − The jack-o-lantern is quite dark, bordering on black, which reduces visual pop
- − Missing the 'twisted trees' in a distinct way compared to the clear silhouettes in model A
Verdict: While FLUX.1 Krea [dev] captures a more haunting atmosphere and beautiful silhouettes, it fails significantly on the text requirements with multiple typos. FLUX.2 [dev] is the superior choice for a functional invitation, as it rendered every requested line of text perfectly while maintaining a high-quality gothic aesthetic.
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.1 Krea [dev]
- + Excellent typography integrated into the scene style
- + Strong lighting and shadow work creating a 3D feel
- + Clean, high-clarity minimalistic aesthetic
- − Flag is interpreted as a physical prop rather than a UI-style icon
- − The salmon texture is slightly over-saturated for a soft cartoon style
FLUX.2 [dev]
- + Features a more complete variety of sushi rolls and nigiri
- + Accurate placement of the flag icon according to prompt instructions
- + Good matte textures on the diorama base
- − Lighting is somewhat flat compared to Model A
- − The text is less stylistically integrated with the 3D scene
Verdict: FLUX.1 Krea [dev] produces a more visually striking image with superior typography and lighting, though it simplifies the subject to one piece of sushi. FLUX.2 [dev] follows the layout instructions more literally, placing the flag as an icon between the text and the diorama, and offers a better variety of food. Overall, FLUX.1 Krea [dev] is the winner due to its professional 3D render quality and cohesive art direction.
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.1 Krea [dev]
- + Excellent sense of motion and 'tumbling' as requested in the prompt
- + Beautifully rendered butterflies with clear wings
- + High-quality fur texture and vibrant lighting
- − Completely missed the 'baby bunny' requirement
- − Generated two kittens instead of one
- − The kitten in the background has slightly unnatural eye alignment
FLUX.2 [dev]
- + Included all requested animals: puppy, kitten, bunnies, and fox
- + Superior lighting with visible god rays and dew sparkles on the grass
- + Higher level of photorealistic detail in the fur and whiskers
- − The animals are sitting relatively still rather than 'chasing and tumbling'
- − The butterflies are less integrated into the action
Verdict: FLUX.2 [dev] is the winner because it successfully included every animal requested in the prompt, including the baby bunny which FLUX.1 Krea [dev] omitted. While FLUX.1 Krea [dev] captured the 'tumbling' action better, FLUX.2 [dev] provided superior 8K-style details, dew effects, and accurate character counts.
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.1 Krea [dev]
- + Excellent hand-drawn etching style that conveys a professional vintage feel.
- + Sophisticated use of texture and decorative borders.
- + Correct date formatting in the requested banner style.
- − Significant spelling error in the primary brand name, rendering it 'FLANOR'IN'.
- − Composition is a bit crowded with the overlapping banner and text.
FLUX.2 [dev]
- + Perfect text rendering with correct spelling and accent mark.
- + Clean, minimalist vector aesthetic that aligns well with modern logo standards.
- + Well-balanced composition with clear hierarchy.
- − The steam icon is a bit generic and simple compared to the vintage prompt.
- − The 'Est. 1720' text is slightly off-center within its banner.
Verdict: While FLUX.1 Krea [dev] produces a beautiful, high-detail illustration with great texture, it fails the basic task of spelling the brand name correctly. FLUX.2 [dev] successfully follows all prompt instructions, including the specific name 'Caffè Florian' and the minimalist vector style, making it the superior choice for a logo.
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.1 Krea [dev]
- + Strong title font and layout
- + Adherence to the color palette
- − Nonsensical iconography including four-person icons and a ringed planet like Saturn
- − Poor adherence to the requested 6-step structure and specific icon descriptions
- − Text rendering is messy and misplaced
FLUX.2 [dev]
- + Excellent adherence to the 6-step timeline and specific icon descriptions
- + Clear, high-quality vector illustrations of the Saturn V and Lunar Module
- + Successfully includes the names of the three astronauts
- − Included extra redundant icons and some garbled text labels
- − Individual icons repeat concepts (e.g., multiple launch/rocket icons)
Verdict: FLUX.2 [dev] significantly outperformed FLUX.1 Krea [dev] by accurately depicting the mechanical details of the Saturn V, Lunar Module, and Moon/Earth orbits as requested. While FLUX.1 Krea produced a more unified composition, the icons themselves were logically incorrect (such as showing Saturn and four astronauts), whereas FLUX.2 adhered closely to the historical and technical requirements of the prompt.
Explore each model
Black Forest Labs' open-weights image generation model with frontier performance, available for non-commercial local deployment