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
HiDream I1 Fast
#51 of 62 in Text-to-Image
Where the votes landed
FLUX.1 [dev]
0%
win rate
Ties
0%
HiDream I1 Fast
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]
- + Excellent photorealism with realistic light refraction and reflections
- + Accurately places the plant behind the cube as requested
- + High detail in the leather texture of the red book
- − The 'cube' is more of a hollow rectangular frame rather than a solid glass cube
HiDream I1 Fast
- + Clean, simple geometric composition
- + Properly follows the spatial instruction of placing the sphere inside and book on top
- − The plant is to the side rather than strictly 'behind' the cube through the glass
- − Lower overall resolution and slightly flatter lighting compared to Model A
- − Reflection on the left side of the glass appears disjointed from the environment
Verdict: FLUX.1 [dev] is the winner due to its superior lighting, texture detail, and adherence to the spatial relationship between the glass cube and the plant in the background. While HiDream I1 Fast produces a clear and acceptable image, it lacks the depth of refraction and the realistic window lighting quality found in FLUX.1 [dev].
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 anatomical accuracy in the hands and facial features.
- + Realistic leather jacket texture and light rain effects.
- + Captured the requested motion blur of passing cars effectively.
- − The man is holding the handlebars rather than actively repairing the bike.
- − The bike appears slightly too modern/sleek for the 'candid street' aesthetic.
HiDream I1 Fast
- + Stronger composition with beautiful reflections on the wet pavement.
- + Better adherence to the 'repairing' aspect of the prompt.
- + Atmospheric lighting and color palette.
- − Significant anatomical errors with the hands, which appear mangled.
- − Structural issues with the bicycle frame near the pedals.
- − Lower overall sharpness and detail compared to the competitor.
Verdict: FLUX.1 [dev] is the winner due to its superior technical execution and realistic skin/clothing textures. While HiDream I1 Fast created a more visually compelling composition with better reflections and repair-oriented posing, it suffered from severe anatomical errors in the hands and a lack of fine detail.
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]
- + Natural skin texture and lighting
- + Captures a subtle, lifelike expression
- + Soft, high-quality depth of field
- − Missed the request for beads in the hair
- − Armor lacks the 'ornate engraving' requested
- − The character looks too clean for 'battle-worn'
HiDream I1 Fast
- + Excellent adherence to the 'beads', 'engraved armor', and 'scars' keywords
- + Strong dynamic lighting from the torches
- + Intricate detail on leather straps and clothing texture
- − Anatomical issues with the hair braids growing directly out of the forehead/scalp
- − Skin texture looks slightly over-sharpened and plastic in some areas
Verdict: While FLUX.1 [dev] produces a more photorealistic face with better skin texture, HiDream I1 Fast followed almost every specific detail of the prompt, including the beads, scars, and engravings. HiDream I1 Fast captures the 'battle-worn' aesthetic much more effectively, even if the composition has some AI-typical hair artifacts.
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 professional layout that truly resembles a real-world restaurant menu.
- + Clean, legible typography and high-quality product photography with a white background.
- + Successful interpretation of 'modern minimalist' aesthetic with well-balanced negative space.
- − Failed to include a specific section for 'Pizza'.
- − Text contains nonsensical placeholder words despite good character formation.
HiDream I1 Fast
- + Successfully used a grid layout for food photos as requested.
- + Attempts to include a header that incorporates words from the prompt.
- − Very low-quality text rendering with illegible, distorted characters.
- − The composition is cramped and lacks the professional design balance of a real menu.
- − Food photos are repetitive and look unappealing due to low resolution and poor lighting.
Verdict: FLUX.1 [dev] produced a highly sophisticated and professional menu design that perfectly captures the requested modern minimalist style and high-quality photography. In contrast, HiDream I1 Fast struggled significantly with text legibility and overall aesthetic appeal, producing a cluttered layout with distorted food images. FLUX.1 [dev] is the clear winner for its superior composition and realistic graphic design quality.
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 photographic quality and lighting
- + Clean explosion layout showing individual layers clearly
- + Accurate rendering of the price '€6.99'
- − Failed to include the main 'MAGIC BURGER' title text
- − Missing the starburst element for the price
- − Background lighting is a bit subdued compared to the 'fiery' request
HiDream I1 Fast
- + Successfully included all requested text elements
- + Dynamic composition with glowing fiery effects and embers
- + Included a starburst for the price as requested
- − Text contains several spelling errors like 'TIMK'
- − Burger is mostly assembled rather than an 'exploded' view of components
- − Lower realism with a more 'digital art' and slightly cluttered feel
Verdict: HiDream I1 Fast followed almost all prompt instructions including the specific text and starburst, but struggled with spelling and the 'exploded' layout. FLUX.1 [dev] produced a much higher quality, more professional image with a true exploded view, but completely omitted the primary branding text 'MAGIC BURGER'. HiDream I1 Fast is the winner for following the complex advertising requirements more closely, despite its lower technical quality.
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 text rendering with near-perfect spelling and alignment.
- + Realistic chalk texture and varying handwriting styles that mimic a human artist.
- + Strong prompt adherence by completing the 'Brown Butter' item logically and including extra relevant text.
- − The title is in all-caps rather than the 'elegant cursive' specifically requested.
- − Minor illegibility in the bottom-most decorative text line.
HiDream I1 Fast
- + Followed the cursive request for the title headline.
- + Pleasant shallow depth-of-field and café atmosphere.
- − Significant text artifacts and overlapping characters rendered items unreadable.
- − Layout is messy with price tags floating and colliding with the text.
- − Failed to maintain a consistent handwriting style across the board.
Verdict: FLUX.1 [dev] is the clear winner as it successfully rendered all requested menu items with high legibility and realistic chalk textures, even completing the truncated prompt for 'Brown Butter' cookies. HiDream I1 Fast struggled significantly with text coherence, resulting in messy overlaps and garbled characters that make the menu unusable.
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]
- + Successfully placed the horse and astronaut in a space setting.
- + Excellent high-detail rendering of the space suit and horse textures.
- + Achieved a cinematic look with dramatic lighting and planet curvature.
- − Failed the specific positional instruction 'horse on top, not vice versa'.
- − The horse's hind legs have anatomical abnormalities in the joint structure.
HiDream I1 Fast
- + Crisp image quality with realistic lighting on the astronaut's gear.
- + Good sense of motion and dust physics in the environment.
- − Completely ignored the 'in space' setting, placing the scene in a desert.
- − Failed the specific positional instruction 'horse on top, not vice versa'.
- − Low creativity relative to the surreal request.
Verdict: Both models failed the specific reasoning challenge of placing the horse on top of the astronaut. However, FLUX.1 [dev] followed the environmental prompt of being 'in space,' whereas HiDream I1 Fast placed the scene in a standard desert. FLUX.1 [dev] is the clear winner for its cinematic atmosphere and adherence to the setting, despite the anatomical errors in the horse's legs.
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]
- + Excellent fur texture and photorealism in the capybara's face.
- + Accurately places the passenger in the back seat as requested.
- + The lighting on the capybara's whiskers and the taxi interior is very natural.
- − The capybara's paws on the steering wheel look slightly distorted.
- − The view out the front windshield is a bit sparse compared to a busy Manhattan street.
HiDream I1 Fast
- + Strong composition showing a side-profile perspective of the taxi interior.
- + Clearer representation of a busy Manhattan street through the window.
- + Correct branding on the taxi driver cap.
- − The capybara has human hands, which is a significant anatomical error.
- − The woman appears to be sitting in the front passenger seat rather than the back seat.
- − The transition between the capybara's head and the jacket looks poorly blended.
Verdict: FLUX.1 [dev] is the clear winner because it correctly places the passenger in the back seat and maintains animal anatomy for the capybara driver. HiDream I1 Fast fails by giving the capybara human hands and placing the passenger in the front seat, which contradicts the prompt's spatial instructions.
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 cinematic lighting with a vibrant center-piece.
- + Intricate thorny border matches the gothic theme well.
- + Clean, readable text for the body and event details.
- − Nonsensical text on the main banner ('Falloween Rantcj').
- − Repetitive time listing and minor typos in the date format.
HiDream I1 Fast
- + Better capture of the 'parchment' aesthetic and traditional invitation layout.
- + Large, readable main title text.
- + Includes both cobwebs and thorny elements in the border decoration.
- − Significant text corruption and overlapping in the footer details.
- − The invitation banner text is cut off and garbled.
- − Lower overall image clarity compared to the competitor.
Verdict: FLUX.1 [dev] produces a much higher quality image with superior lighting and composition, though it fails significantly on the primary header text. HiDream I1 Fast better interprets the 'parchment' and 'webs' instructions but suffers from severe text artifacts and overlapping lines in the lower half of the invitation. FLUX.1 [dev] is the likely winner for its polish and legibility of the core event details, despite the gibberish banner.
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 adherence to the 'miniature 3D cartoon scene' aesthetic with soft, clay-like textures.
- + Very balanced isometric composition and colors.
- + Includes a flag icon as requested.
- − Text rendering is poor, resulting in 'SUSH CATON' and random characters.
- − The diorama base is very simple and lacks the 'raised' wooden look typically associated with sushi presentation.
HiDream I1 Fast
- + Perfect text rendering for both 'JAPAN' and 'SUSHI'.
- + Accurately depicts specific sushi types like nigiri and gunkan-maki on a traditional wooden base.
- + Strict adherence to the 45-degree isometric perspective.
- − The nigiri has a strange cross-section graphic on the side of the rice.
- − Lighting is a bit flat compared to the soft, realistic PBR materials requested.
Verdict: HiDream I1 Fast is the superior choice for this prompt because it followed the text instructions perfectly, whereas FLUX.1 [dev] failed significantly on text rendering. While FLUX.1 [dev] captured the 'soft refined texture' slightly better, the overall clarity and accuracy of the objects and typography in HiDream I1 Fast make it much more usable.
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 unavailable for this challenge.
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]
- + Excellent vector aesthetic with a clean professional balance.
- + Accurately includes the steam and complex banner requested.
- + High-quality subtle texture on the background.
- − Significant spelling errors in both the main brand name and the word 'restaurant'.
- − Includes extraneous numbers like '11011' and '1941' that weren't requested.
HiDream I1 Fast
- + Correctly spelled the primary brand name 'Caffè Florian'.
- + Strong use of the requesting warm brown and cream color palette.
- + Clearer minimalist composition focusing on the cloche dome.
- − Added an extra digit to the year, making it '17210'.
- − The text placement inside the dome is a bit cramped and visually busy.
- − The steam is rendered as abstract circles rather than natural vapor.
Verdict: HiDream I1 Fast is the winner because it successfully spelled the core brand name correctly, whereas FLUX.1 [dev] produced 'Caffé Flariláan'. Although both models struggled with number accuracy (FLUX.1 added extra dates and HiDream I1 Fast added a digit to the year), HiDream I1 Fast captured the requested minimalist emblem style more effectively for an actual logo design.
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]
- + Excellent adherence to the color palette and flat-vector style requirements.
- + Clean, professional composition with smooth lines and sophisticated layout.
- + High visual quality with a consistent design language across icons and typography.
- − Text is largely gibberish despite the professional appearance.
- − Iconography is abstract and doesn't clearly map to all specific steps requested (e.g., Saturn V vs generic rocket).
HiDream I1 Fast
- + Attempts to label the specific steps requested in the prompt.
- + Contains a relevant flag icon for the lunar landing phase.
- + Follows the NASA-inspired color palette effectively.
- − Visual quality is poor with jagged edges and artifacts.
- − The layout is messy with random lines connecting icons that don't follow a logical mission path.
- − Text is misspelled and suffers from significant distortion.
Verdict: FLUX.1 [dev] is the clear winner despite its inability to render the specific requested text accurately. It successfully captures the 'clean, modern vector' aesthetic with a highly professional layout and crisp graphics. HiDream I1 Fast fails on the technical execution, producing an image with low-resolution artifacts, messy line work, and distorted text.
Explore each model
Distilled version of HiDream AI's 17B parameter text-to-image model