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
Recraft V4.1
#53 of 62 in Text-to-Image
Where the votes landed
FLUX.2 [klein] 9B
0.0%
win rate
Ties
0.0%
Recraft V4.1
100.0%
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 adherence to lighting instructions with a clear window light source from the left.
- + Highly realistic glass texture with appropriate thickness and internal reflections.
- + Natural plant integration behind the glass.
- − The glass object is more of a hollow vase/container than a solid cube.
- − The perspective of the table surface is slightly warped on the right side.
Recraft V4.1
- + Perfect geometric cube shape as requested.
- + Very sharp, clean focus on the central subject.
- + Beautifully rendered wood grain and reflections on the table.
- − The plant is positioned to the side rather than 'behind the cube partially visible through the glass'.
- − The sphere appears to be floating mid-air inside a solid block without any physical support or hollow space logic.
Verdict: FLUX.2 [klein] 9B followed the spatial instructions more accurately by placing the plant behind the glass, creating realistic refraction. Recraft V4.1 produced a more aesthetically pleasing 'cube', but failed on the specific positional requirement for the plant and created a more surrealistic floating effect.
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 anatomical detail in the face and hands including realistic skin texture.
- + Superior rendering of the bicycle mechanics and the tool kit in the foreground.
- + Effective use of shallow depth of field reflecting a 50mm lens look.
- − The cars in the background lack the 'motion blur' requested.
- − The lighting on the man feels slightly staged rather than purely natural street light.
Recraft V4.1
- + Successfully captured the 'motion blur' from passing cars and a more 'imperfect framing' look.
- + Atmospheric rendering of rain and wet pavement reflections.
- + Authentic candid street photography mood.
- − The man's posture and anatomy are slightly off, with his head positioned strangely relative to his body.
- − Less detail in the bicycle and tools compared to the other model.
- − The skin texture and facial details are obscured by shadow and lower resolution.
Verdict: FLUX.2 [klein] 9B produces an image with much higher technical clarity and better anatomical accuracy, particularly in the face and hands of the man. However, Recraft V4.1 better followed the specific stylistic cues like 'motion blur' and 'imperfect framing', creating a moodier, more candid atmosphere despite its lower detail levels.
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 adherence to the 'beads in hair' prompt with clearly visible details.
- + Superior rendering of the 'leather straps and cloth underlayer' textures.
- + Striking facial expressions and lifelike eye detail that convey the 'battle-worn' theme.
- − The lighting feels a bit more staged/artificial compared to the more cinematic Model B.
- − The 'scars' look slightly painted on rather than integrated into the skin texture.
Recraft V4.1
- + Strong dynamic composition with a more cinematic, natural-looking torchlight effect.
- + Excellent shallow depth of field and bokeh sparks that frame the subject well.
- + Realistic integration of dirt and grime onto the skin surface.
- − The hair beads are much less distinct and plentiful than requested.
- − The texture on the leather straps is less defined compared to Model A.
Verdict: FLUX.2 [klein] 9B followed the prompt details more accurately, particularly regarding the hair beads and the specific textures of the leather and cloth. However, Recraft V4.1 produced a more atmospheric and cinematic image with superior lighting and a more natural 'battle-worn' appearance, making it the preferred choice for visual storytelling despite missing small decorative details.
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
- + Clean layout with vibrant image grid
- + Successfully includes all requested headers: Appetizers, Pizza, Mains
- − Tex rendering is unintelligible gibberish
- − The food images are repetitive, mostly showing different pizzas for every section
- − Layout has overlapping text elements at the top
Recraft V4.1
- + Perfect text legibility with realistic menu items and descriptions
- + Excellent professional layout that mirrors a real menu design
- + Diverse and high-quality food photography that correlates with the text
- − The 'grid' format of photos is more of a vertical strip than a traditional grid
- − Minimalist style might be considered slightly plain by some
Verdict: Recraft V4.1 is the clear winner as it produces an actual functional menu with perfectly legible English text and relevant dish photography. While FLUX.2 [klein] 9B achieves a nice aesthetic balance, its text is completely unreadable and it fails to vary the food types, showing pizza in every header category.
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
- + Exceptional photorealistic rendering of food textures like grilled meat and bun.
- + Professional typography integration with high-quality glowing effects.
- + Dynamic and balanced composition that feels like a real commercial advertisement.
- − The burger is stacked closely rather than fully 'exploded' as requested in the prompt.
Recraft V4.1
- + Accurately interprets the 'exploded' request with clear separation between all layers.
- + Highly creative and stylised fiery text for the main title.
- + Good sense of motion with the sauce splashes and drifting smoke.
- − The starburst element looks like a low-quality clip-art sticker rather than a glowing effect.
- − Overall food photography looks less appetizing and slightly flatter than the competitor.
Verdict: FLUX.2 [klein] 9B produces a much more polished and professional-looking advertisement with superior lighting and texture detail, though it played it safe with the 'exploded' layout. Recraft V4.1 followed the spatial instructions of the prompt more literally and featured impressive main titles, but it failed significantly on the starburst element and overall image cohesion.
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 chalk texture including realistic eraser smudges and dust.
- + Beautiful cursive header that matches the aesthetic requirement.
- + Very high legibility and consistent cursive style throughout.
- − Contains a spelling error in the footer ('fress' instead of 'fresh').
- − Finished the 'Brown Butter' item with 'Cookies' which wasn't fully in the prompt but made a logical leap.
Recraft V4.1
- + Perfect text accuracy, correctly completing the 'Brown Butter' item.
- + Highly realistic, imperfect handwriting that feels genuinely manual.
- + Consistent chalk texture across all items.
- − The header is in print rather than the requested elegant cursive.
- − The top edge of the board is cut off in the frame.
Verdict: Recraft V4.1 followed the specific text instructions more accurately, including correctly finishing the 'Brown Butter' line without making assumptions, though it failed to use cursive for the title. FLUX.2 [klein] 9B produced a more visually pleasing composition with a beautiful cursive title and better framing, but it suffered from a typo in the footer. Recraft V4.1 is the winner for its superior adherence to the handwriting style and perfect spelling.
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
- + Excellent high-detail rendering of the spacesuit and planets
- + Cinematic composition with diverse celestial bodies
- + Clear and sharp focus on the subjects
- − Nonsensical text watermarked at the bottom of the image
- − Failed the spatial counter-instruction; the astronaut is riding the horse
Recraft V4.1
- + Dynamic lighting with a strong backlit effect on the horse's mane
- + More realistic integration of the subjects into the space environment with floating debris
- + Good textures on the horse and asteroids
- − Anatomical issues with the horse's front and rear legs
- − Failed the spatial counter-instruction; the astronaut is riding the horse
Verdict: Both models failed the negative constraint to put the horse on top of the astronaut, instead interpreting the prompt as a standard 'astronaut riding a horse.' FLUX.2 [klein] 9B produced a cleaner, more colorful cinematic image but suffered from distracting AI-generated text, while Recraft V4.1 offered better lighting and environmental depth despite some anatomical warping on the horse.
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
- + Perfectly places the passenger in the back seat as requested
- + Includes specific details like the dark jacket and professional capybara expression
- + Excellent rendering of the businesswoman's bored expression and the phone lighting on her face
- − Composition feels slightly cramped from the front dashboard perspective
- − The capybara's front paws look more like gloves than paws
Recraft V4.1
- + Side-profile composition clearly shows both the capybara and the passenger in their designated seats
- + The capybara's paws on the steering wheel look more realistic
- + Good cinematic lighting and texture on the capybara's fur
- − The capybara is not wearing the requested dark jacket
- − The capybara is positioned very high, seemingly floating in the seat
- − The businesswoman's posture looking at the phone is slightly awkward
Verdict: FLUX.2 [klein] 9B adhered more strictly to the prompt by including the jacket and capturing the specific 'bored' expression of the passenger, though the front-facing viewpoint makes the spatial arrangement feel tight. Recraft V4.1 provided a more cinematic side-on view that clearly distinguishes front and back seats, but it missed the jacket requirement and the passenger's face is less expressive. FLUX.2 [klein] 9B is the winner for its superior attention to the character details and atmospheric lighting.
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
- + Excellent text legibility and alignment throughout the image.
- + Perfect adherence to all prompt elements including the scroll banner and specific event details.
- + Clean, polished composition with high-quality cinematic lighting.
- − The jack-o-lantern and trees look a bit generic and digital compared to Model B.
Recraft V4.1
- + Atmospheric and gritty texture that fits the 'vintage gothic' request perfectly.
- + Intricate hand-drawn feel to the typography and border details.
- + Highly expressive and detailed jack-o-lantern.
- − The scroll banner text is too small and difficult to read.
- − Composition of bottom text is slightly cluttered and less 'invitation-like' than Model A.
Verdict: Both models performed exceptionally well, but FLUX.2 [klein] 9B is the clear winner for a functional invitation as it perfectly balanced aesthetic with highly legible, accurate text for all requested details. While Recraft V4.1 had superior vintage textures and artistic flair, its tiny banner text and scattered layout made the secondary information hard to process.
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 typography rendering for 'JAPAN' and 'SUSHI'.
- + Clean isometric representation with smooth textures.
- + Perfect soft lighting and shadow on the diorama base.
- − The flag icon is incorrect, resembling the flag of Yemen rather than Japan.
- − The sushi rolls have slightly generic 3D asset textures.
Recraft V4.1
- + Features the correct national flag of Japan.
- + Highly realistic PBR materials, especially on the fish textures and rice grains.
- + Includes logical extras like chopsticks that add to the diorama feel.
- − The text 'JAPAN' and 'SUSHI' is smaller and less bold than requested.
- − The '45° top-down' angle is slightly too low compared to a true isometric view.
Verdict: While FLUX.2 [klein] 9B delivers better typography and a truer isometric perspective, it fails on the specific detail of the Japanese flag. Recraft V4.1 produces a much higher quality of materials and correct iconography, making it the more representative and visually impressive image overall.
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 color vibrancy and lighting composition.
- + Whimsical and clean artistic style that fits the joyful vibe.
- − Completely missed the baby bunny from the prompt.
- − The kitten's pose and eyes look slightly unnatural/AI-generated.
Recraft V4.1
- + Higher prompt adherence by including all four animals: puppy, kitten, fox, and bunny.
- + Very realistic fur textures and interaction between the animals.
- + Subtle and professional god rays and dew sparkle effects.
- − The puppy's mouth has some slightly messy teeth rendering.
Verdict: Recraft V4.1 is the clear winner as it successfully included all four requested animals, including the baby bunny which FLUX.2 [klein] 9B omitted. Additionally, Recraft V4.1 achieved a more realistic 'hyper-photorealistic' look with detailed fur and natural interactions, whereas FLUX.2 [klein] 9B leaned more towards a stylized, illustrative 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.2 [klein] 9B
- + Excellent adherence to the 'vector emblem' and 'banner' prompt elements
- + Strong typography that is well-balanced with the icon
- + Beautifully rendered subtle texture on the background
- − The steam element looks slightly like a solid hair-swirl rather than vapor
Recraft V4.1
- + Elegant high-contrast vector style with clean lines
- + Accurate representation of steam above the cloche
- + Precise typography and accent mark placement
- − Misses the 'subtle texture' on the background
- − The logo elements feel a bit disconnected compared to the circular emblem in Model A
Verdict: Both models followed the prompt exceptionally well, producing clean and professional logos. FLUX.2 [klein] 9B is preferred for its superior composition, creating a cohesive circular emblem that feels like a complete brand package, whereas Recraft V4.1 produced a more fragmented layout on a flat white background.
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
- + Strong visual hierarchy with integrated background elements
- + Follows the NASA-inspired color palette effectively
- + Creative use of silhouettes for crew members
- − Abysmal text rendering with numerous spelling errors in every label
- − Logical flow is confusing and non-linear
- − Incorrect icons used (e.g., full moon for Descent step)
Recraft V4.1
- + Excellent text legibility and perfect spelling of key mission stages
- + Consistent flat-vector iconography that adheres strictly to the prompt
- + Clear, logical grid layout that follows the specified steps chronologically
- − Numbering error in the steps (repeats '04')
- − Missed the light gray background, optng for a solid navy background instead
- − Iconography is very simple, bordering on generic
Verdict: Recraft V4.1 is the clear winner as it produces a functional infographic with perfect spelling and a logical grid layout. While FLUX.2 [klein] 9B has a more ambitious and textured artistic style, its catastrophic failure to spell basic words like 'Launch' and 'Landing' makes it unusable for an infographic task.
Explore each model
Recraft's V4.1 standard tier text-to-image model — refines V4's photorealism with more natural lighting, softer gradients, and sharper illustration styles for everyday creative work