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
Settled by community votes across 16 shared challenges, with an AI judge weighing in on each.
FLUX.2 [pro]
#8 of 62 in Text-to-Image
HiDream I1 Full
#60 of 62 in Text-to-Image
Where the votes landed
FLUX.2 [pro]
0%
win rate
Ties
0%
HiDream I1 Full
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 [pro]
- + Excellent photographic realism and lighting
- + Clean, sharp glass rendering with accurate reflections
- + Perfect adherence to all spatial instructions
- − The sphere is resting on the bottom rather than floating (though not specified as floating, it creates a more static feel)
HiDream I1 Full
- + Natural look to the glass material
- + Good capture of the plant behind the cube
- − Significant ghosting artifacts with the blue sphere appearing three times
- − Lighting on the table is a bit harsh compared to 'soft' window light requested
- − Distorted circular table edge versus the clean perspective in Model A
Verdict: FLUX.2 [pro] produced a much higher quality image with superior clarity and realistic textures. HiDream I1 Full failed on the details of the 'single' blue sphere, creating multiple ghosted versions of it within the cube and suffering from some perspective distortions.
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 [pro]
- + Excellent adherence to the 'imperfect framing' and 'candid' prompts by positioning the subjects to the side.
- + Very realistic light rain effects and interactions with the pavement.
- + Anatomically correct interaction with the bicycle components (repairing the chain/pedal area).
- − The motion blur on the passing cars is subtle, though present.
- − Slightly less 'natural skin texture' detail compared to the other model.
HiDream I1 Full
- + Strong reflections on the wet pavement.
- + Good central focus with high contrast.
- + Satisfies the 'elderly Japanese man' prompt well with high facial detail.
- − The man appears to be sitting on or straddling the bicycle rather than repairing it.
- − The bicycle anatomy is broken, with the man's leg passing through the frame of the bike.
- − Lacks the 'imperfect framing' requested, opting for a standard centered composition.
Verdict: FLUX.2 [pro] followed the complex instructions much better, specifically capturing the 'candid' and 'imperfect framing' aspects that typical AI models often ignore in favor of centered subjects. While HiDream I1 Full has high-quality textures, it suffers from significant spatial clipping where the man's leg passes through the bicycle frame, whereas FLUX.2 [pro] maintains realistic physical interaction.
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 [pro]
- + Excellent photorealistic skin texture and lifelike eyes.
- + Natural, high-quality engraving on the plate armor with realistic lighting physics.
- + Detailed texture on the worn leather straps and coarse cloth underlayer.
- − The beads in the hair look slightly like modern plastic craft beads rather than period-appropriate materials.
HiDream I1 Full
- + Strong adherence to the 'battle-worn' prompt with visible blood-tinged scars.
- + Good symmetrical composition and clear engraving on the armor.
- − The skin texture appears overly smoothed or 'AI-filtered' compared to Model A.
- − Lighting on the face is inconsistent with the light sources in the background.
Verdict: FLUX.2 [pro] is the clear winner due to its superior realism, particularly in the rendering of skin pores, moisture in the eyes, and the physical material properties of the armor and leather. While HiDream I1 Full captures the 'battle-worn' aesthetic well, its overall image quality feels more like a digital illustration than a lifelike photograph.
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 [pro]
- + Excellent typography with legible bold sans-serif fonts.
- + Sophisticated grid layout that correctly separates all requested sections.
- + High-quality, varied food photography that looks professional.
- − Includes some spelling errors like 'MINS' instead of 'MAINS'.
- − Repeats 'Margherita Pizza' multiple times in the list view.
HiDream I1 Full
- + Features a clean, minimalist aesthetic with ample whitespace.
- + Produces high-quality, vibrant food photography.
- + Good use of bold headers for clear navigation.
- − All food photos are of the same dish (pizza), ignoring appetizers and other mains.
- − The layout is less structured, with text overlapping the white background inconsistently.
- − Text contains significant gibberish and lacks proper currency symbols/formatting.
Verdict: FLUX.2 [pro] is the clear winner as it adheres much better to the grid layout and varied section requirements of the prompt. While HiDream I1 Full produces high-quality images, it failed to show variety in the food photos and had inferior text rendering compared to FLUX.2 [pro].
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 [pro]
- + Excellent adherence to the menu items and prices requested in the prompt.
- + Highly realistic chalk texture with genuine-looking hand-drawn variations and smudges.
- + Perfect spelling and coherent typography for the entire text block.
- − The 'Today's Specials' title is not in all-caps as specifically requested by the prompt.
HiDream I1 Full
- + Features a nice wooden frame and an illustrative coffee cup drawing.
- + The date text is clear and placed prominently.
- − Fails completely on menu content, producing gibberish text instead of the requested dishes.
- − The text looks more like a digital chalk font rather than unique handwriting.
- − Major errors in text rendering, including repeated '$50' prices and nonsensical words.
Verdict: FLUX.2 [pro] followed the prompt's content requirements almost perfectly, rendering the specific menu items with high legibility and realistic chalk textures. In contrast, HiDream I1 Full failed to produce the requested menu items, providing a board full of illegible gibberish and digital-looking fonts. FLUX.2 [pro] is the clear winner for its superior language processing and realistic artistic execution.
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 [pro]
- + Excellent adherence to the 'horse on top' spatial instruction
- + Highly detailed rendering of both the spacesuit and the horses
- + Strong cinematic lighting and composition with high surrealism
- − Anatomical glitches where the horse's legs clip through the astronaut
HiDream I1 Full
- + Dynamic sense of movement and speed
- + Clean lighting and clear rendering of the astronaut and horse
- − Failed the core prompt instruction of placing the horse on top
- − Common interpretation of astronaut on horse, lacking surrealism
- − Lower resolution/clipping in the background atmosphere
Verdict: FLUX.2 [pro] correctly interpreted the complex and counter-intuitive spatial instruction of a 'horse riding an astronaut,' capturing the surreal nature requested. While it has some minor clipping artifacts, it is vastly superior to HiDream I1 Full, which defaulted to a standard astronaut-on-horse image, ignoring the specific prompt constraints.
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 [pro]
- + Excellent photorealistic lighting and textures
- + Incredibly detailed car interior with accurate dashboard components
- + Perfect adherence to 'paws on the steering wheel' and 'taxi driver cap' style
- − The passenger is slightly less detailed compared to the main subject
HiDream I1 Full
- + Clear facial expressions on both the capybara and the human
- + Strong color saturation and contrast
- − The capybara's 'hands' look more like human-monkey hybrids than capybara paws
- − The cap is a modern baseball cap rather than a traditional taxi driver flat cap
- − Interior lacks the grit and detail of a real NYC taxi
Verdict: FLUX.2 [pro] wins decisively due to its superior photorealism, capturing the gritty atmosphere of a New York taxi at night with realistic window rain and dashboard details. While HiDream I1 Full produces a clear image, it lacks the professional anatomical accuracy in the capybara's paws and the nuanced lighting that makes FLUX.2 [pro] look like an actual photograph.
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 [pro]
- + Perfect text rendering for all requested details, including date and location.
- + Excellent atmospheric lighting and a more sophisticated gothic aesthetic.
- + Superior texture on the parchment and jack-o-lantern.
- − The parchment borders are slightly cut off at the edges of the frame.
HiDream I1 Full
- + Good inclusion of standard Halloween motifs like webs and bats.
- + Strong contrast in the central pumpkin illustration.
- − Text rendering is poor with gibberish words like 'Yootmber' and 'Attep'.
- − Failed to include the specific date, time, and location details requested.
- − The illustration style is more cartoonish than 'polished/cinematic'.
Verdict: FLUX.2 [pro] is the clear winner as it followed every instruction, including the complex text requirements for dates and addresses, while maintaining a high-quality cinematic atmosphere. HiDream I1 Full failed significantly on text accuracy and provided a more generic, lower-quality illustration style.
Bald man challenge
Image Editing“Give the person a full, thick head of natural hair with realistic texture, density, and a natural hairline. Preserve facial features and lighting.”
AI Judge Analysis
FLUX.2 [pro]
- + Successfully added thick, realistic hair with a natural hairline.
- + Excellent preservation of the source background, clothing, and original facial features.
- + The lighting on the hair matches the sunlight direction in the scene.
- − The forehead skin texture looks slightly smoothed compared to the source image.
HiDream I1 Full
- + Maintained the general aesthetic of the jacket.
- − Failed the primary instruction by leaving the top of the head bald.
- − Added bizarre hallucinated objects including a strap across the face and a backpack.
- − Completely changed the background landscape and the man's facial features.
Verdict: FLUX.2 [pro] followed the editing instructions perfectly, providing a realistic head of hair while keeping the identity of the person and the environment intact. In contrast, HiDream I1 Full failed to add hair on top, significantly altered the person's face, changed the background, and added nonsensical artifacts to the subject's face.
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 [pro]
- + Perfectly follows the text instructions, including the small flag icon.
- + Superior rendering of textures, particularly the subsurface scattering on the ginger and fish.
- + Clean, professional typography and layout that feels like a deliberate graphic design.
- − The rice texture on the front nigiri is slightly repetitive and bead-like.
HiDream I1 Full
- + Appealing 3D cartoon style with good variety in the sushi types.
- + Accurate isometric perspective and grounding on the diorama base.
- − Completely missed the 'small flag icon' requested in the prompt.
- − The 'SUSHI' text is too small and lacks the 'large bold' presence requested compared to 'JAPAN'.
- − Visible artifacts on the left edge of the image and less refined material rendering.
Verdict: FLUX.2 [pro] is the clear winner as it followed every detail of the prompt, including the specific text layout and the flag icon which HiDream I1 Full omitted. FLUX.2 [pro] also produced much higher quality materials and lighting, resulting in a cleaner and more professional-looking isometric diorama.
Over-the-top cartoon caricature
Editing“Create a caricature of me and my job. Make it exaggerated and humorous, incorporating my profession as a tv show anchor and my love for dogs and hockey.”
AI Judge Analysis
FLUX.2 [pro]
- + Excellent adherence to all prompt requirements including the anchor job, dogs, and hockey.
- + Clear caricature style that captures the subject's features in an exaggerated way.
- + Highly creative composition with dogs as co-anchors and a themed news desk.
- − The facial resemblance to the source image is slightly more generalized into a cartoon style.
HiDream I1 Full
- + Strong preservation of the source subject's facial features while applying a caricature effect.
- + Good inclusion of the dog element in the foreground.
- − Fails to include any hockey elements as requested.
- − The 'TV anchor' profession is only vaguely implied by a TV in the background or a small headset on the dog.
- − The composition is less creative compared to Model A.
Verdict: FLUX.2 [pro] followed the complex instructions much better than HiDream I1 Full, successfully integrating the hockey, TV anchor, and dog elements into a cohesive and humorous caricature. HiDream I1 Full preserved the subject's likeness more accurately but failed to include the hockey request and had a much weaker interpretation of the professional setting.
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 [pro]
- + Excellent adherence to the prompt by including all four distinct animals: puppy, kitten, bunny, and fox kit.
- + Superior photorealistic quality with highly detailed fur textures and realistic anatomy.
- + Effective lighting with soft god rays and visible dew sparkles on the grass.
- − The interaction is slightly static rather than 'playfully chasing' and 'tumbling'.
HiDream I1 Full
- + High contrast and vibrant colors that contribute to a joyful, wholesome vibe.
- + Clear, expressive eyes on the animals as requested.
- − Missing the baby bunny entirely, failing one part of the prompt.
- − Appears less like a photograph and more like a digital illustration or 'AI-stylized' image.
- − Anatomical issues, particularly the fox's front right leg which appears disjointed.
Verdict: FLUX.2 [pro] followed the prompt much more accurately, successfully depicting all four requested animals, whereas HiDream I1 Full missed the rabbit and included two kittens instead. Furthermore, FLUX.2 [pro] achieved a much higher level of photorealistic detail and realistic lighting, whereas HiDream I1 Full had a more saturated, digital art appearance with anatomical flaws.
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 [pro]
- + Excellent preservation of the original pose, character placement, and clothing styles.
- + Captures the wash-style watercolor and pencil-line aesthetic of Studio Ghibli backgrounds.
- + Perfectly translates the iconic facial expressions from the 'Distracted Boyfriend' meme into an anime style.
- − The colors are slightly more muted/faded than typical modern Ghibli films.
HiDream I1 Full
- + Vibrant color palette and cute character designs.
- + Good interpretation of 'dreamy backgrounds' with a town-like feel.
- − Poor source preservation; changed the man's pose, the woman's hair, and added a beard.
- − The guy is no longer looking back at the woman in red, losing the core context of the meme image.
- − The style feels more like generic modern anime than the specific hand-painted Ghibli aesthetic.
Verdict: FLUX.2 [pro] is the clear winner as it successfully maintains the exact composition and narrative of the source image while perfectly applying the requested Studio Ghibli art style. In contrast, HiDream I1 Full loses the essential 'distracted' pose of the original meme and replaces the characters with generic avatars that don't resemble the source subjects.
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 [pro]
- + Excellent source preservation, maintaining the woman's face and the dog perfectly.
- + Strong adherence to all parts of the prompt, including blowing hair and flying leaves.
- + Consistent visual quality with the source image.
HiDream I1 Full
- + Successfully captures a highly energetic and lively feeling through body pose.
- + Effective hair motion and leaf placement.
- − Poor source preservation as the dog is completely removed from the scene.
- − Significant изменения to the woman's facial features and clothing, making it look like a different person.
- − The background and overall environment were heavily altered.
Verdict: FLUX.2 [pro] is the clear winner as it successfully follows the edit instructions while keeping the integrity of the original photo intact, including the person's identity and the dog. In contrast, HiDream I1 Full failed the basic requirement of image editing by removing major elements of the source (the dog) and completely changing the subject's face.
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 [pro]
- + Excellent font rendering with accurate spelling and accents.
- + High-quality vector illustration style with sophisticated shading and texture.
- + Superior composition that feels like a completed professional logo.
- − The 'Est. 1720' text is slightly small compared to the main brand name.
HiDream I1 Full
- + Simplified minimalist vector shapes.
- + Accurate adherence to requested color palette.
- − Completely failed to include the brand name 'Caffè Florian'.
- − The cloche looks more like a bell or a heavy dome with confusing internal coffee cup iconography.
- − The banner edges are disconnected from the central design.
Verdict: FLUX.2 [pro] is the clear winner as it followed all prompt instructions, including the specific brand name and accurate 'Est. 1720' banner. HiDream I1 Full failed to include the primary text 'Caffè Florian' and produced a disjointed composition with less refined vector paths.
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 [pro]
- + Excellent layout that logically flows through the mission steps.
- + High-quality vector aesthetic with clean iconography and professional font rendering.
- + Follows the NASA-inspired color palette perfectly.
- − The placeholder body text is gibberish.
- − The Saturn V rocket icon is replaced by a generic shuttle-style rocket.
HiDream I1 Full
- + Bold, attractive title treatment for 'APOLLO 11'.
- + Includes a more recognizable Saturn V-style rocket silhouette.
- − Fails to include more than half of the requested mission steps.
- − Text rendering is messy with nonsensical additions like '+ UNG' and '+ ICON'.
- − Layout is cluttered and lacks a logical chronological flow.
Verdict: FLUX.2 [pro] followed the complex instructions much more effectively, producing a complete multi-step infographic that maps the mission from launch to landing. While HiDream I1 Full has a strong graphic title, it failed to include several prompt requirements and the layout is disorganized compared to the cohesive design of FLUX.2 [pro].
Explore each model
HiDream AI's 17B parameter text-to-image model using sparse diffusion transformer with mixture of experts, achieving state-of-the-art image generation quality with strong prompt following