Black Forest Labs' compact, open-source image generation model with sub-second inference, optimized for production and near real-time applications with multi-reference support
Settled by community votes across 13 shared challenges, with an AI judge weighing in on each.
FLUX.2 [klein] 4B
#32 of 62 in Text-to-Image
HiDream I1 Fast
#50 of 62 in Text-to-Image
Where the votes landed
FLUX.2 [klein] 4B
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] 4B
- + Excellent adherence to the 'small' scale for the sphere.
- + Realistic wood grain and book texture.
- + Accurate lighting and reflections on the bottom surface of the cube.
- − The glass cube walls appear slightly inconsistent in thickness.
- − The plant in the background is less sharp compared to the foreground objects.
HiDream I1 Fast
- + Clean glass edges and very clear rendering of the cube.
- + Soft window lighting is handled well.
- − Failed the scale requirement; the blue sphere is large, nearly filling the cube.
- − The sphere appears to be floating rather than resting on the bottom surface.
- − The red book has a somewhat plastic, non-paper texture.
Verdict: FLUX.2 [klein] 4B is the superior image as it correctly interpreted the spatial and scale instructions, keeping the sphere 'small' within the cube. HiDream I1 Fast produced a visually clean image, but the oversized, floating sphere and less realistic book texture made it less faithful to the prompt.
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] 4B
- + Excellent depiction of light rain and atmospheric moisture
- + High level of detail in skin texture and clothing materials
- + Accurate representation of a 50mm lens perspective with natural shadows
- − Physical logic errors with the bicycle frame intersecting the man's leg
- − Lack of significant motion blur on background cars as requested
HiDream I1 Fast
- + Effective use of reflections on the wet pavement
- + Good cinematic composition and color palette
- + Subject pose feels natural for the scene
- − Lower resolution and noticeable softness in textures
- − The man appears to be sitting in the air rather than on the bicycle or a seat
- − Fails to show visible rain, appearing more like post-rain dampness
Verdict: FLUX.2 [klein] 4B produces a much more detailed and realistic image with superior skin textures and convincing rain effects, though it has some structural clipping issues with the bicycle. HiDream I1 Fast captures a moody atmosphere but suffers from lower image quality and a glaring anatomical/structural error where the man is floating above the bike. FLUX.2 is the clear winner for its adherence to the 'natural skin texture' and 'light rain' prompts.
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] 4B
- + Excellent photorealistic skin texture and lifelike eyes.
- + Subtle and realistic battle damage including faint scars and dirt.
- + Ornate engraving on the armor is clear and consistent with the metal surface.
- − The braids and beads are a bit simple compared to the prompt's potential.
- − The background fire looks slightly artificial.
HiDream I1 Fast
- + Intricate braiding and beadwork that highly matches the prompt.
- + Dramatic and vibrant warm torchlight lighting and bokeh effects.
- + Great detail on leather straps and armor ornaments.
- − The scars look like red paint or makeup rather than real tissue.
- − The image has a slightly 'digital painting' look that reduces the requested lifelike realism.
- − The anatomy of the hair braids merging into the armor is slightly confusing.
Verdict: FLUX.2 [klein] 4B produces a much more realistic portrait with superior skin textures and convincing battle-worn details. While HiDream I1 Fast excels at the decorative elements like beads and braids, it fails to achieve the 'lifelike' quality of the eyes and skin seen in the FLUX image.
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] 4B
- + Excellent photographic quality and food styling
- + High aesthetic coherence with a clean minimalist design
- + Better use of grid-based layout for multiple sections
- − Some gibberish text in headings like 'Mains' spelled as 'Mdains'
- − Text rendering for smaller menu items is messy
HiDream I1 Fast
- + Bold headers and vibrant red accents
- + Attempted to group sections more clearly
- + Clearer numbers for pricing
- − Low resolution and grainy image artifacts
- − Poor food photo quality compared to the other model
- − Significant spelling errors in the Title
Verdict: FLUX.2 [klein] 4B produces a much more professional and visually appealing layout with high-quality photography that looks like a real restaurant menu. While both models struggle with typography, HiDream I1 Fast has significant image quality issues, including low resolution and poor food rendering.
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] 4B
- + Excellent photorealistic texture on the bun and patties.
- + Includes all requested text elements clearly, including the starburst price tag.
- + High-quality lighting and particle effects create a strong sense of motion.
- − The main title 'MAGIC BURGER' is significantly misspelled with overlapping letters.
- − The burger is stacked rather than being truly 'exploded' with components in mid-air.
HiDream I1 Fast
- + Correct spelling of the main 'MAGIC BURGER' title.
- + Includes more diverse ingredient elements floating around the main burger.
- + Vibrant fiery background with actual glowing coals at the bottom.
- − Features repetitive and messy text, including a redundant price stamp on the right.
- − Photorealism is lower with some plastic-looking ingredient textures.
- − Includes strange donut-like rings that were not in the prompt.
Verdict: While FLUX.2 [klein] 4B produces a much more professional-looking image with superior photorealistic textures, it fails significantly on the spelling of the main title. HiDream I1 Fast manages to spell the main title correctly and captures more of the 'exploded' feel, but suffers from cluttered secondary text and lower visual fidelity. FLUX.2 [klein] 4B is preferred for its overall aesthetic and successful rendering of the starburst and secondary text, despite the garbled main title.
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] 4B
- + Excellent chalk texture and realistic smudge marks on the board.
- + Consistently elegant cursive handwriting that matches the prompt's style request.
- + Includes completing the truncated prompt with relevant text and prices.
- − Multiple spelling errors across almost every word (e.g., 'Truffel', 'Musheram', 'Ootprous').
- − Spacing issues in the title with 'S TPECIALS'.
HiDream I1 Fast
- + Beautiful environmental composition and lighting within a café setting.
- + Better initial spelling for the main headers and menu items.
- − Significant text artifacts and overlapping characters in the pricing area.
- − Illegible scribbles and ghost text appear between the menu lines.
- − Fails to maintain a consistent chalk texture, looking more like digital overlays in some areas.
Verdict: FLUX.2 [klein] 4B is the better choice for this prompt as it successfully captures the specific 'chalk texture' and consistent handwriting style requested, despite its poor spelling. HiDream I1 Fast creates a more visually appealing scene overall, but the text itself is plagued by messy artifacts, overlapping numbers, and digital-looking render errors that ruin the 'handwritten chalkboard' requirement.
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] 4B
- + Successfully places the scene in space as requested
- + High level of detail in the space suit and horse's coat
- + Correctly interprets the 'horse on top' instruction relative to the planet
- − Anatomical issues with the horse's legs and hooves resting on thin air
- − The shadow on the planet's atmosphere is physically impossible
HiDream I1 Fast
- + Dynamic sense of motion with the horse's gallop
- + Reasonable lighting on the space suit
- − Completely fails the 'in space' prompt, placing the scene in a desert
- − Distorted horse anatomy around the legs and hooves
- − Low resolution with visible compression artifacts
Verdict: FLUX.2 [klein] 4B followed the prompt much more effectively by actually placing the subject in space with a cinematic background. HiDream I1 Fast ignored the 'in space' requirement, depicting an astronaut in a desert with significantly lower image quality and clarity.
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] 4B
- + Excellent photorealism in lighting and textures.
- + Perfect adherence to the 'both front paws on the steering wheel' instruction.
- + The businesswoman accurately captures a bored, normal expression in the backseat.
- − The capybaras ear is awkwardly clipping through the hat.
- − The car interior proportions feel a bit cramped for a standard sedan.
HiDream I1 Fast
- + Strong bokeh effect and vibrant city lights.
- + Clear 'TAXI' text on the driver's hat.
- − The capybara has human hands instead of paws as requested.
- − The passenger is shown in the front seat or a mid-cabin position rather than clearly in the back seat.
- − The composition feels less like a real photo and more like a digital composite.
Verdict: FLUX.2 [klein] 4B is the clear winner for its superior photorealism and strict adherence to specific prompts, particularly the capybara's paws on the wheel and the passenger's placement. HiDream I1 Fast fails significantly by giving the animal human hands and misplacing the passenger, resulting in a less coherent scene.
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] 4B
- + High visual quality with realistic lighting and texture
- + Excellent composition and gothic atmosphere
- + Includes all requested decorative elements like thorns and webs correctly
- − Significant spelling errors in the main title text
- − Missing the 'Time' and 'Year' digits in the bottom text section
HiDream I1 Fast
- + Closer adherence to the parchment poster layout requested
- + Correctly spelled the words 'Halloween Party' at the top
- + Included the full set of specified event details at the bottom
- − Illustration style is flat and cartoonish compared to the cinematic request
- − Text in the scroll and the location details are badly garbled
- − Lower overall resolution and clarity
Verdict: While FLUX.2 [klein] 4B produces a much more atmospheric and high-quality image, it fails significantly on text legibility and accuracy. HiDream I1 Fast succeeds in creating the requested 'parchment poster' look and spells the title correctly, but suffers from messy text at the bottom and a less sophisticated art style. FLUX is preferred for its superior aesthetic value even though both models struggled with complex text instructions.
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] 4B
- + Excellent PBR material rendering on the salmon and rice
- + Clean, high-resolution aesthetic with professional lighting
- + Strong adherence to the 45-degree isometric perspective request
- − Typos in the text with 'SUSH' instead of 'SUSHI'
- − Includes an incorrect flag icon (resembles Austria/Latvia rather than Japan)
HiDream I1 Fast
- + Accurately renders all requested text including 'JAPAN' and 'SUSHI'
- + Includes a correct Japanese flag icon
- + Better interpretation of the 'raised diorama base' with the wooden platform
- − The sushi models are less realistic and more plasticky
- − Visual artifacts such as the floating line and distorted icon next to 'SUSHI'
- − Lower overall clarity and texture refinement compared to model A
Verdict: While FLUX.2 [klein] 4B produces a much more visually stunning image with superior textures and lighting, it fails on basic text and iconography accuracy. HiDream I1 Fast successfully follows the complex text and icon instructions but suffers from lower image quality and several graphical artifacts. HiDream I1 Fast is likely preferred here purely for following the specific labeling instructions, despite the lower artistic quality.
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] 4B
- + Successfully includes the god rays and dew sparkles requested in the prompt.
- + Each animal's fur texture is very distinct and sharp.
- + The composition feels dynamic and accurately reflects the 'chasing' and 'tumbling' motion.
- − Failed to include the baby bunny, providing two kittens instead.
- − The paw placement on the standing kitten looks slightly detached from its body.
HiDream I1 Fast
- + The lighting is very warm and captures the 'wholesome vibe' well.
- + Face details on the puppy and fox are very expressive and cute.
- + High visual quality with a soft, dream-like bokeh effect.
- − Failed to include the baby bunny, providing two kittens instead.
- − The scene is static rather than showing the animals 'tumbling together' or 'chasing' butterflies.
Verdict: Both models failed to include all four requested animals, omitting the bunny in favor of a second kitten. FLUX.2 [klein] 4B is the better choice as it successfully rendered specific atmospheric details like god rays and captured a sense of movement, whereas HiDream I1 Fast produced a more static, posed portrait.
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] 4B
- + Clean vector aesthetic suitable for a real logo
- + Accurate establishment date rendering
- + Excellent background paper texture
- − Failed to spell the name correctly (FLAXTION instead of Florian)
- − Redundant text repetition with the date appearing twice
HiDream I1 Fast
- + Perfectly spelled the main brand name 'Caffè Florian'
- + Warm and inviting gold and brown color palette
- + Good use of steam and celestial decorative elements
- − Typos in the establishment year ('17210' instead of '1720')
- − Text placement inside the cloche feels cramped and less like a professional logo
Verdict: HiDream I1 Fast captures the correct branding name and a more classic vintage feel, although it fails on the year. FLUX.2 [klein] 4B produces a much cleaner vector-style image that feels like a professional logo, but the misspelling of 'Florian' and the redundant text layers make it less usable without editing.
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] 4B
- + Excellent vector art style with clean lines and subtle gradients.
- + Consistent iconography and a color palette that matches the NASA-inspired request.
- + Includes specific mission details like the names Aldrin and Collins and the Tranquility landing site.
- − Severely garbled main title text.
- − Iconography for the steps is confusing and doesn't clearly follow the 1-6 sequence.
HiDream I1 Fast
- + Main title 'APOLLO 11' is clearly legible and well-rendered.
- + Better layout flow for an infographic, following the central rocket.
- + More legible labels for the individual mission steps.
- − Visual style is messy with jagged lines and 'leaking' colors, failing the 'crisp lines' requirement.
- − Icons are nonsensical (e.g., Saturn-like rings for Translunar, a flag for Lunar Orbit).
- − Poor execution of the requested flat-vector aesthetic.
Verdict: FLUX.2 [klein] 4B followed the stylistic requirements perfectly, producing a professional-looking vector illustration, though it failed significantly on text legibility. HiDream I1 Fast achieved better text for the title and labels, but the actual illustration quality is very low, with messy lines and nonsensical icons that do not resemble the professional infographic style requested. FLUX.2 [klein] 4B is the better choice as its visual assets are usable for a design project even if the text needs replacing.
Explore each model
Distilled version of HiDream AI's 17B parameter text-to-image model