Black Forest Labs' 12-billion parameter flow transformer for high-quality text-to-image generation, suitable for personal and commercial use with streaming support
Settled by community votes across 13 shared challenges, with an AI judge weighing in on each.
FLUX.1 [dev]
#16 of 62 in Text-to-Image
FLUX.2 [pro]
#8 of 62 in Text-to-Image
Where the votes landed
FLUX.1 [dev]
0%
win rate
Ties
0%
FLUX.2 [pro]
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 [dev]
- + Exceptional photographic realism in materials
- + High-quality refractions and reflections within the glass
- + Satisfying color saturation
- − The sphere appears to be floating unnaturally or fused with the glass
- − The cube looks more like a solid glass block with a cavity rather than a hollow cube
FLUX.2 [pro]
- + Perfect adherence to the physical relationship of the objects
- + The plant is clearly visible through the glass as requested
- + More realistic wood grain and soft lighting
- − The blue sphere is quite small relative to the cube
- − Slightly more muted colors compared to image A
Verdict: Both models followed the complex spatial instructions perfectly. FLUX.1 [dev] produced a more striking, high-contrast image with beautiful glass textures, but FLUX.2 [pro] followed the instruction for the plant being visible through the glass more effectively and maintained a more convincing physical composition.
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 [dev]
- + Excellent shallow depth of field with realistic background bokeh.
- + Very high-quality clothing texture and leather jacket detail.
- + Captures a realistic 'candid' moment with the man's posture.
- − The man appears to be just holding the bike rather than actively repairing it.
- − The motion blur of the passing car is very subtle compared to the request.
FLUX.2 [pro]
- + Strong composition with better adherence to the 'repairing' aspect of the prompt.
- + Highly realistic wet pavement reflections including ripple effects in puddles.
- + Successful inclusion of motion blur in the background traffic.
- − The bike anatomy becomes slightly garbled near the chainring and pedals.
- − Slightly less 'natural' skin texture compared to Model A.
Verdict: FLUX.2 [pro] followed the prompt more comprehensively by showing the man actively kneeling to repair the bike and clearly incorporating motion blur from passing cars. While FLUX.1 [dev] produced a cleaner image with superior clothing textures, it felt more like a posed portrait rather than a scene of a man performing a repair task.
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 [dev]
- + Features very lifelike, piercing eyes that align with the prompt.
- + Captures a subtle bokeh effect with soft lighting.
- − Missed several specific prompt elements like beads in the hair and faint scars.
- − The character looks too clean and 'pristine' for a battle-worn description.
FLUX.2 [pro]
- + Excellent adherence to specific details like hair beads, scars, and dirt.
- + Highly detailed texture on leather straps, cloth underlayers, and engraved armor.
- + Stronger interpretation of the torchlight and battle-worn aesthetic.
- − The sparks look slightly more like brush strokes than photographic bokeh compared to the other model.
Verdict: FLUX.2 [pro] followed the prompt with significantly higher accuracy, including all requested details like the hair beads, scars, and specific textures. While FLUX.1 [dev] produced a high-quality portrait, it ignored the 'battle-worn' request, resulting in a character that looks too tidy for the description.
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 [dev]
- + Excellent minimalist graphic design aesthetic
- + Correct use of bold sans-serif fonts
- + Clean professional layout with balanced white space
- − Failed to include a grid of colorful food photos
- − Text becomes garbled nonsense in the descriptions
- − Only includes two categories (Appetizers/Mains) instead of the three requested
FLUX.2 [pro]
- + Successfully implemented the food photo grid with vibrant colors
- + Followed the prompt instructions for specific categories (Appetizers, Pizza, Mains)
- + Legible header text and price formatting
- − Confusing repetition of categories in the layout
- − Misspelled section header 'MINS' instead of 'Mains'
- − Image contents are repetitive (listing Garlic Bread under every section)
Verdict: FLUX.2 [pro] is the stronger choice for this specific prompt as it correctly followed the instructions for a 'food photo grid' and specific menu categories, whereas FLUX.1 [dev] produced a more abstract design that ignored the grid requirement. While FLUX.1 [dev] has a more sophisticated aesthetic, FLUX.2 [pro] captures the functional requirements of a casual dining menu more effectively.
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 [dev]
- + Excellent photorealistic rendering of the food items.
- + Strong vertical composition that looks professional and clean.
- + Good use of focal depth and lighting.
- − Failed to include the 'MAGIC BURGER' title text.
- − The background effect is minimal and lacks the 'fiery, glowing' text effect requested.
- − The price is not in a starburst shape.
FLUX.2 [pro]
- + Perfect adherence to text prompts, including the title, secondary message, and starburst price.
- + Highly dynamic composition with sauce droplets and crumbs conveying motion.
- + Beautifully rendered 'fiery' text effect as requested.
- − The bun texture is slightly less realistic compared to Model A.
Verdict: FLUX.2 [pro] followed the prompt instructions near-perfectly, successfully integrating all requested text elements with the specific fiery stylistic requirements and the starburst shape. While FLUX.1 [dev] produced a very attractive food image, it completely missed the main title, making it a failed advertisement relative to the specific prompt.
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 [dev]
- + Excellent legibility and clean character formation.
- + Accurately rendered all requested menu items and prices.
- + Effective use of an actual wooden frame for the chalkboard.
- − The text looks a bit too much like a clean digital font rather than authentic dusty chalk.
- − The handwriting style for the title is blocky rather than the requested 'elegant cursive'.
FLUX.2 [pro]
- + Superb realism in the chalk texture, including dust and varying pressure.
- + Followed the instruction for 'elegant cursive' handwriting for the title.
- + The slate material has a very realistic, textured edge.
- − Text alignment is slightly sloping and crowded toward the bottom.
- − Some characters have slight artifacts, making them a bit harder to read than Model A.
Verdict: FLUX.1 [dev] produced a very clean and readable menu with perfect prompt adherence for text content, but it lacked the gritty, authentic chalk texture requested. FLUX.2 [pro] excelled at the stylistic requirements, providing a much more convincing 'hand-drawn' look with beautiful cursive and realistic chalk dust, despite slightly less orderly layout.
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 [dev]
- + Clean, cinematic lighting with a strong focal point.
- + High anatomical clarity of the horse and suit textures.
- − Fails the specific prompt instruction to have the horse on top of the astronaut.
FLUX.2 [pro]
- + Successfully follows the difficult constraint of having the horse on top of the astronaut.
- + Very high cinematic detail in the space background and nebulae.
- − Anatomical issues with the rear horse sharing legs or merging into the astronaut's back.
- − Slightly cluttered composition with the multiple galaxies.
Verdict: While FLUX.1 [dev] produced a more visually polished and aesthetically pleasing image, it completely ignored the specific spatial instruction 'horse on top'. FLUX.2 [pro] followed the logic of the prompt, creating a surreal image where the horse is literally riding/standing on the astronaut, making it the winner for prompt adherence.
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 [dev]
- + Successfully positions the woman in the back seat as requested
- + Rendered high-quality fur and whisker details on the capybara
- + Captures the bored expression on the passenger perfectly
- − The steering wheel is on the wrong side for a New York taxi (Right Hand Drive)
- − The woman appears to be sitting in the front passenger seat due to the perspective of the car frame
FLUX.2 [pro]
- + Executes a much more complex composition with the passenger clearly in the background
- + Excellent photorealistic details in the taxi dashboard and rainy window effects
- + Follows the jacket and cap descriptions with high fidelity
- − The capybara's hands look slightly more human/gloved than natural paws
- − The steering wheel orientation is still ambiguous relative to the frame
Verdict: FLUX.2 [pro] is the superior image due to its much more convincing composition and detailed environment, successfully placing the passenger in the back seat while maintaining focus on the capybara driver. FLUX.1 [dev] failed the spatial logic of the prompt, placing the passenger in the front seat right next to the driver.
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 [dev]
- + Strong illustrative style with a clean thorn border.
- + Vibrant glowing light effect from the jack-o-lantern.
- + Includes the requested scroll banner element.
- − Significant spelling errors in the title text ('Falloween Ranty', 'You Tre').
- − Redundant time entry ('7pm, 7pm') and inconsistent punctuation in the date.
- − Art style feels more like a modern vector illustration than a 'vintage gothic parchment'.
FLUX.2 [pro]
- + Perfect text rendering for all requested details including title, banner, and event info.
- + Excellent adherence to the 'vintage gothic' and 'dark parchment' aesthetic.
- + Higher realism in the central jack-o-lantern and moody atmosphere.
- − The thorns and webs are combined into a darker, less distinct border compared to Model A.
- − Lighting on the scroll feels slightly detached from the foggy environment.
Verdict: FLUX.2 [pro] is the clear winner as it successfully rendered all the complex text requirements perfectly and accurately captured the vintage dark parchment aesthetic. FLUX.1 [dev] struggled significantly with text legibility and opted for a more cartoonish illustrative style that didn't match the requested gothic atmosphere as well.
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 [dev]
- + Excellent 3D miniature 'clay-like' texture
- + Soft, gentle lighting that creates a peaceful atmosphere
- + Accurate 45-degree isometric perspective
- − Text contains typos with 'SUSH CATON' and a garbled character
- − The diorama base and text are quite small relative to the canvas
- − Text and flag are floating semi-randomly in the background
FLUX.2 [pro]
- + Perfect text rendering for both 'JAPAN' and 'SUSHI'
- + Strong adherence to the 'bold text at top-center' instruction
- + Clean, high-clarity 3D modeling and lighting
- − The sushi rice texture looks a bit like uniform bubbles/cubes rather than realistic grain
- − The background blue is a bit darker than requested but still acceptable
Verdict: FLUX.2 [pro] followed the layout and text instructions much better than FLUX.1 [dev], delivering perfectly rendered typography and a cleaner composition. While FLUX.1 [dev] had a slightly more charming 'miniature' texture, its failure to render the requested text accurately makes FLUX.2 [pro] the superior result for this specific prompt.
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 [dev]
- + Features a very clean and vibrant color palette with warm golden lighting.
- + Captures the 'wholesome' and 'joyful' vibe effectively with stylized expressions.
- − Failed the prompt requirement for realism, resulting in a 3D animation/cartoon style.
- − Missing the tabby kitten requested in the prompt.
- − Anatomical issues with paws and ears, appearing more like plush toys than animals.
FLUX.2 [pro]
- + Perfectly adheres to the 'hyper-photorealistic' instruction with realistic fur textures and anatomy.
- + Correctly includes all four requested animals: golden retriever, tabby kitten, bunny, and fox kit.
- + Excellent integration of butterflies and dew sparkles in a natural-looking meadow.
- − The lighting, while realistic, is slightly less dramatic than the 'god rays' requested in the prompt.
Verdict: FLUX.1 [dev] produced a stylized, cartoonish image that failed to include the kitten and ignored the request for photorealism. FLUX.2 [pro] followed every detail of the prompt accurately, delivering a highly realistic scene with all four animals and impressive anatomical detail. FLUX.2 [pro] is the clear winner for its superior prompt adherence and visual quality.
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 [dev]
- + Strong vintage aesthetic with woodblock-style cross-hatching
- + Good use of warm brown tones
- − Major spelling errors in the brand name ('Flarilaan') and category ('Reseaurant')
- − Addition of random, hallucinated numbers '11011' and '1941'
- − Cloche illustration is very basic and small relative to the text
FLUX.2 [pro]
- + Perfect text rendering for both the main name and the foundation date
- + More sophisticated vector emblem style with professional shading
- + Subtle paper texture matches the prompt requirements perfectly
- − The cloche handle is slightly off-center within the steam trail
Verdict: FLUX.2 [pro] followed the prompt accurately, producing a clean, professional logo with perfect spelling and high-quality vector details. In contrast, FLUX.1 [dev] failed significantly on text rendering, hallucinating extra numbers and misspelling the primary restaurant name.
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 [dev]
- + Captures the requested navy, muted red, and white palette well.
- + Clean flat-vector aesthetic with a balanced, circular composition.
- − The text is largely nonsensical gibberish.
- − Includes random planets like Saturn that were not part of the prompt.
- − Fails to follow the specific 6-step logical sequence requested.
FLUX.2 [pro]
- + Excellent text rendering for headings and crew names.
- + Strictly follows the requested steps from Launch to Landing in a logical flow.
- + High-quality iconography that matches the Apollo mission details.
- − Missed the 'subtle gradients only' instruction by using some heavier shading on the moon.
- − The 'Translunar' step iconography is a bit cluttered with multiple Earths.
Verdict: FLUX.2 [pro] is the clear winner as it successfully rendered the specific 6-step sequence and readable text requested in the prompt. While FLUX.1 [dev] produced a pleasant aesthetic, it failed on the technical details of the infographic, including gibberish text and irrelevant planetary icons.
Explore each model
Black Forest Labs' state-of-the-art image generation model with maximum quality and speed, supporting text-to-image and multi-reference image editing with up to 4MP output