Black Forest Labs' distilled 9 billion parameter image generation model with sub-second inference and multi-reference support
Settled by community votes across 13 shared challenges, with an AI judge weighing in on each.
FLUX.2 [klein] 9B
#13 of 32 in Image Editing
HiDream I1 Fast
#51 of 62 in Text-to-Image
Where the votes landed
FLUX.2 [klein] 9B
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.2 [klein] 9B
- + Excellent photorealism and natural lighting effects.
- + Highly detailed wood texture and glass reflections.
- + Accurate interpretation of 'small blue sphere' relative to the cube size.
- − The glass object is more of a rounded vessel than a strict geometric cube.
- − The sphere appears to be levitating without a clear physical contact point.
HiDream I1 Fast
- + Perfect geometric cube shape with thin glass edges.
- + Good alignment with all prompt elements including placement of the book and plant.
- + Clean, minimalist composition.
- − The blue sphere is quite large, bordering on 'medium' rather than 'small'.
- − Lighting on the table feels a bit flat compared to Model A.
- − The glass edges and sphere contact points look slightly artificial.
Verdict: FLUX.2 [klein] 9B wins on pure visual quality and realism, capturing the soft window light and wood textures beautifully, though its cube is somewhat rounded. HiDream I1 Fast followed the geometric instructions better with a sharp-edged cube, but the overall image feels more like a 3D render than a real photograph. FLUX.2 is preferred for its superior lighting and material detail.
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 [klein] 9B
- + Excellent skin texture and hyper-realistic facial details.
- + Highly accurate interaction between the subject, his tools, and the bicycle components.
- + Superior technical execution of reflections, rain particles, and depth of field.
- − The car in the background lacks the requested motion blur, appearing mostly static.
HiDream I1 Fast
- + Captures a wider city street atmosphere with good reflections.
- + Includes blurred cars in the background as requested in the prompt.
- − The subject appears to be sitting awkwardly on a very small child-sized bicycle rather than repairing it.
- − Poor anatomical and object consistency, such as the man's hands merging into the frame.
- − Visible AI artifacts and a lack of 'natural skin texture' compared to the other model.
Verdict: FLUX.2 [klein] 9B is the clear winner due to its exceptional realism and precise adherence to the activity of 'repairing' the bike, whereas HiDream I1 Fast depicts a man awkwardly perched on a disproportionately small bicycle. FLUX.2 provides tangible textures and believable physics, making it a much more convincing candid street photo.
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 [klein] 9B
- + Excellent realization of 'highly detailed texture' on the leather straps and chainmail/cloth underlayers.
- + Highly realistic skin textures and scars that feel integrated into the face.
- + Superior rendering of the ornate engraved plate armor with natural light reflections.
- − The torch in the background is a bit distracting in its placement relative to the head.
HiDream I1 Fast
- + Strong composition with a dynamic three-quarters pose.
- + Colorful beads in the hair add a nice visual variety.
- + Good use of bokeh and sparks to create atmosphere.
- − The textures on the straps and skin look smoother and less 'battle-worn' than requested.
- − The lighting on the face feels a bit artificial compared to the environment.
- − The scar on the forehead looks more like a surface painting than a physical mark.
Verdict: FLUX.2 [klein] 9B provides a superior interpretation of the 'battle-worn' request, featuring incredibly detailed textures on the leather and skin that look much more realistic. While HiDream I1 Fast has a more colorful and cinematic aesthetic, it lacks the fine tactile detail on the armor and clothing that the prompt specifically requested.
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 [klein] 9B
- + Exhibits a very clean, professional minimalist layout typical of modern graphic design.
- + Correctly includes the requested headers (Appetizers, Pizza, Mains) in a bold sans-serif font.
- + The image quality of the food photography is high and well-lit.
HiDream I1 Fast
- + Successfully follows the grid layout requested in the prompt.
- + Colors are vibrant and match the casual dining aesthetic.
- − Text rendering is very poor and contains significant gibberish.
- − The header uses a black background, which contradicts the 'white background' requirement.
- − Food images are repetitive and lower in relative quality.
Verdict: FLUX.2 [klein] 9B followed the prompt much more successfully, delivering a professional-grade layout that actually looks like a real menu. HiDream I1 Fast failed on the specific background color request and produced text that was significantly more distorted. FLUX.2's integration of the headers and the clean white space makes it the superior choice for a design-related prompt.
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 [klein] 9B
- + Excellent typography with perfect spelling and high-quality fiery effects.
- + Highly photorealistic textures on the bun, patties, and vegetables.
- + Superior composition that feels like a professional advertising poster.
- − The 'exploded' effect is a bit conservative, with layers still mostly stacked.
HiDream I1 Fast
- + Creative use of glowing coals in the foreground to establish the theme.
- + Dynamic inclusion of floating ingredients like onion rings.
- − The secondary text is garbled and unreadable.
- − The starburst design and overall lighting feel more like a flat graphic than a photorealistic 3D scene.
- − Lower resolution details on the patty and vegetables compared to the competitor.
Verdict: FLUX.2 [klein] 9B followed every instruction perfectly, producing a professional-grade advertisement with crisp, accurate text and beautiful photorealistic lighting. HiDream I1 Fast struggled significantly with text rendering and the overall quality of the burger textures, resulting in a less polished image.
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 [klein] 9B
- + Excellent typography with perfect spelling following the prompt's cut-off text.
- + Realistic chalk texture including smudges and dust on the board.
- + Consistent handwriting style that truly looks like chalk rather than a digital font.
- − Small typo in the bottom extra text ('fress' instead of 'fresh').
HiDream I1 Fast
- + Provides environmental context with a cozy café setting.
- + Includes the requested 'April 30, 2026' date accurately.
- − Numerous spelling errors and text overlapping (e.g., 'and Her4s', 'Buter').
- − Inconsistent price placement and messy chalk artifacts that look like glitches.
- − Illegible words and messy layout compared to a real menu.
Verdict: FLUX.2 [klein] 9B is the clear winner for its superior text rendering, maintaining perfect spelling and a highly realistic chalk texture that matches the prompt's aesthetic. HiDream I1 Fast struggles significantly with text legibility, containing several overlapping characters and misspellings that make the menu unreadable.
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 [klein] 9B
- + Successfully depicts the horse and astronaut in a space environment.
- + High level of detail in the textures of the horse's coat and the spacesuit.
- + Dynamic composition with numerous celestial bodies and a cinematic feel.
- − Failed the negative constraint; the horse is being ridden by the astronaut instead of being on top.
- − Includes nonsensical AI-generated text at the bottom of the image.
HiDream I1 Fast
- + Good lighting on the subject and realistic dust effects around the hooves.
- + Clean rendering of the spacesuit and horse tack.
- − Failed the primary prompt location; depicted a desert on Earth rather than space.
- − Failed the negative constraint; the horse is being ridden by the astronaut instead of being on top.
- − Composition is relatively generic compared to the requested surreal style.
Verdict: Both models failed the specific 'horse on top' spatial reasoning request, defaulting to the common trope of an astronaut riding a horse. FLUX.2 [klein] 9B is the superior image because it correctly placed the subjects in space with high detail, whereas HiDream I1 Fast ignored the space setting entirely, placing the scene in a terrestrial desert.
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 [klein] 9B
- + Excellent photorealism in textures, particularly the capybara's fur and the woman's coat.
- + Accurate composition with the passenger correctly placed in the back seat.
- + Coherent interior lighting that matches the phone screen and city lights.
- − The text on the cap is slightly garbled ('NEW TALA').
- − The capybara's paws look somewhat like human hands in gloves.
HiDream I1 Fast
- + Clear text on the cap ('TAXI').
- + Captures the 'bored expression' of the woman well.
- − The passenger is sitting in the front passenger seat instead of the back seat as requested.
- − The capybara has distinct human hands clearly visible on the steering wheel, which is jarring.
- − The composition feels flatter and less integrated into a realistic car environment.
Verdict: FLUX.2 [klein] 9B is the clear winner as it correctly followed the spatial instruction to place the passenger in the back seat, whereas HiDream I1 Fast placed her in the front next to the driver. FLUX.2 also achieved a much higher level of photoregularism and better anatomical integration, despite the minor text error on the hat.
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 [klein] 9B
- + Perfect text rendering for all requested titles, slogans, and event details.
- + Excellent composition with sophisticated cinematic lighting and atmospheric depth.
- + Integrated the border elements (webs and thorns) seamlessly into a high-quality vintage aesthetic.
- − The jack-o-lantern is quite traditional and could have a more 'gothic' twist to match the font.
HiDream I1 Fast
- + Contains all the requested elements like the parchment, pumpkin, and bats.
- + Clear, bold colors that pop against the dark background.
- − Significant text errors including '7:m e con', 'Anches, NYC', and garbled text on the banner.
- − Flat, cartoonish illustration style that lacks the requested 'cinematic' and 'polished' feel.
- − Lower overall resolution and visual coherence compared to the competitor.
Verdict: FLUX.2 [klein] 9B is the clear winner as it followed every instruction perfectly, including complex text rendering for dates and locations without a single typo. While HiDream I1 Fast included the general elements, it suffered from severe text artifacts and a much more simplistic, clip-art style of illustration.
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 [klein] 9B
- + Excellent 3D texture rendering with beautiful soft lighting.
- + Clean and stylish typography for the labels.
- + Strong isometric perspective that fits the prompt perfectly.
- − The flag icon is incorrect, resembling the flag of Yemen or the German Empire rather than Japan.
- − The sushi preparation is slightly confusing, with a nigiri-style topping on a maki-style roll.
HiDream I1 Fast
- + Successfully included the correct Japanese flag icon.
- + Higher contrast and vibrant colors that pop against the background.
- + Includes additional thematic elements like chopsticks.
- − The perspective is more eye-level than the requested 45-degree top-down isometric view.
- − Text layout is a bit cluttered with extra lines and icons.
- − There is a logical error in the sushi where a filling is visible on the side of the nigiri rice.
Verdict: Both models followed the prompt well, but FLUX.2 [klein] 9B produced a superior 3D aesthetic with more refined textures and a perfect isometric angle. While HiDream I1 Fast correctly rendered the Japanese flag, its composition was less balanced and failed to achieve the specific top-down perspective requested.
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 [klein] 9B
- + Excellent adherence to the lighting prompt with clear god rays and dew sparkles.
- + High fur detail and sharp texture rendering across all subjects.
- + Highly expressive and playful posing of the animals.
- − Missing the baby bunny requested in the prompt.
- − The kitten's pose and anatomy look slightly unnatural/distorted under the puppy's paw.
HiDream I1 Fast
- + Features a soft, warm color palette that fits the 'wholesome' vibe.
- + Good fur texture on the foxes and kittens.
- + Accurate eye expressions for the requested 'big expressive eyes'.
- − Missing the baby bunny, and instead included two kittens.
- − The resolution and sharpness are significantly lower than the competitor.
- − Does not show the animals 'tumbling together' as they are mostly sitting still.
Verdict: Both models failed to include all four requested animals, specifically omitting the baby bunny. FLUX.2 [klein] 9B is the superior image due to its much higher resolution, better execution of the 'god rays' lighting, and more dynamic, playful composition that better fits the description of animals tumbling in a meadow. HiDream I1 Fast appears blurry in comparison and failed to capture the requested action.
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 [klein] 9B
- + Excellent typography with correct spelling and accent mark.
- + Clean vector-style execution with high professional polish.
- + Accurate rendering of the requested date '1720' on the banner.
- − The steam element is a bit thick and singular compared to traditional steam wisps.
HiDream I1 Fast
- + Creative placement of the main text inside the cloche.
- + Good color palette adherence.
- − Typographical error in the date: '17210' instead of '1720'.
- − Line work is messy with disconnected circular borders and artifacts around the steam.
- − Text inside the cloche is slightly misaligned.
Verdict: FLUX.2 [klein] 9B is the clear winner as it produced a professional, production-ready logo with perfect spelling and clean vector lines. HiDream I1 Fast failed on the specific text requirements, adding an extra digit to the date and showing significant rendering artifacts in the linework.
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 [klein] 9B
- + Stronger adherence to the flat-vector style with crisp, clean lines.
- + The composition follows a logical flow that tells a story of the mission.
- + Includes silhouettes of the three crew members which adds a creative touch.
- − Numerous spelling errors including 'AOILLO', 'EARDHT', and 'LANDINING'.
- − The Saturn V rocket design is a bit generic and lacks the iconic black/white pattern.
HiDream I1 Fast
- + Successfully spelled 'APOLLO 11' correctly in the main header.
- + The central rocket icon is more recognizable as a Saturn V variant.
- + Closely followed the requested NASA-inspired color palette.
- − The layout is cluttered and the connecting lines are messy and nonsensical.
- − The text for individual steps is heavily garbled and unreadable.
- − Includes an orange planet with a ring that looks more like Saturn than a 'Translunar' step.
Verdict: FLUX.2 [klein] 9B followed the infographic prompt much better, creating a logical flow of information with high-quality vector aesthetics, despite several typos in the labels. HiDream I1 Fast struggled with the structure of the infographic, resulting in a disorganized layout and extremely poor text rendering for the sub-steps.
Explore each model
Distilled version of HiDream AI's 17B parameter text-to-image model