Head to head
Esc

Models · slot A

to navigate to pick

FLUX.2 [klein] 4B Black Forest Labs HiDream I1 Full HiDream AI

Settled by community votes across 16 shared challenges, with an AI judge weighing in on each.

FLUX.2 [klein] 4B

22.3 arena score

#32 of 62 in Text-to-Image

Skill signature · Text-to-Image

HiDream I1 Full

13.8 arena score

#60 of 62 in Text-to-Image

Vote tally

Where the votes landed

FLUX.2 [klein] 4B

0.0%

win rate

Ties

0.0%

HiDream I1 Full

100.0%

win rate

0.0% 0.0% ties 100.0%
Shared challenges 16

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.”

FLUX.2 [klein] 4B
HiDream I1 Full

AI Judge Analysis

FLUX.2 [klein] 4B

  • + Excellent reflection of the blue sphere on the bottom glass panel.
  • + Very clean and sharp rendering of the book texture.
  • + Strong spatial logic between the objects.
  • The glass cube is missing a top panel, appearing more like an open box.

HiDream I1 Full

  • + Successfully depicts the plant through the glass as requested.
  • + Good lighting atmosphere with natural-looking shadows.
  • + Properly enclosed cube shape.
  • The blue sphere is levitating rather than sitting inside the cube.
  • Strange artifacts/ghosting reflections of the blue sphere on the side panels that don't match physics.

Verdict: FLUX.2 [klein] 4B followed the prompt with high clarity and realistic textures, though it rendered the cube as an open container. HiDream I1 Full captured the refractive quality of the glass and the background plant better, but failed on physical logic with a floating sphere and messy reflections. FLUX.2 is the winner for its superior sharpness and material realism.

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.”

FLUX.2 [klein] 4B
HiDream I1 Full
0% wins 0% ties 100% wins

AI Judge Analysis

FLUX.2 [klein] 4B

  • + Excellent depiction of light rain and atmospheric depth.
  • + High level of detail in the bicycle components and textures.
  • + Realistic skin texture and natural pose.
  • The rear bicycle tire is floating off the ground and lacks a logical connection to the frame.
  • The car movement is static rather than showing the requested motion blur.

HiDream I1 Full

  • + Beautiful reflections on the wet pavement.
  • + Good central composition and color grading.
  • + Captures the 'candid' feel of the street scene effectively.
  • Significant anatomical and mechanical errors: the man has three legs.
  • The bicycle is tiny and out of scale compared to the man.
  • Lacks the motion blur from cars requested in the prompt.

Verdict: While both models failed to include the requested motion blur from passing cars, FLUX.2 [klein] 4B is the clear winner for its superior realism and technical coherence. HiDream I1 Full suffered from Hallucinating a third leg on the subject and severe scaling issues with the bicycle, making it an inferior image despite the nice environmental reflections.

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.”

FLUX.2 [klein] 4B
HiDream I1 Full

AI Judge Analysis

FLUX.2 [klein] 4B

  • + Excellent photorealism in the skin texture and facial features
  • + Clean and intricate engraving details on the plate armor
  • + Subtle and realistic dirt/scarring that feels integrated into the skin
  • The hair braids are relatively simple compared to the prompt's potential for complexity
  • Composition is a bit generic and centered

HiDream I1 Full

  • + Strong adherence to hair beads and complex braiding
  • + Intense and high-contrast lighting creates a more dramatic atmosphere
  • + Detailed leather straps and varied armor textures
  • Over-sharpened and slightly 'plastic' skin texture
  • Visual artifacts such as a floating strap end and messy hair-to-armor transitions
  • The sparks and bokeh look a bit artificial compared to the subject

Verdict: FLUX.2 [klein] 4B produces a much more realistic and believable portrait with skin textures that look truly human, though it is conservative with the hair details. HiDream I1 Full captures more of the specific decorative elements like beads and complex straps but suffers from over-processing and digital artifacts. FLUX.2 is the superior choice for overall visual quality and coherence.

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.”

FLUX.2 [klein] 4B
HiDream I1 Full

AI Judge Analysis

FLUX.2 [klein] 4B

  • + Excellent grid layout with varied food photography types.
  • + Sophisticated typography that mimics a real menu structure.
  • + Better adherence to the 'vibrant accents' and 'professional layout' requirements.
  • Text is largely gibberish upon close inspection.
  • The section headers are somewhat repetitive or misspelled ('MDAINS', 'MDANS').

HiDream I1 Full

  • + Text is relatively clear and readable.
  • + Simple, clean layout that is easy to follow.
  • Lacks variety in food photos, showing only pizzas for different categories.
  • Composition feels a bit basic and less professional for a 'modern minimalist' design.
  • Top margin is cut off.

Verdict: FLUX.2 [klein] 4B produces a much more realistic and professionally designed menu layout with a sophisticated grid of varied food items. While HiDream I1 Full provides clearer text, it fails on variety by showing only pizzas across all sections, making it less useful for the 'appetizers/mains' prompt requirement.

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.”

FLUX.2 [klein] 4B
HiDream I1 Full

AI Judge Analysis

FLUX.2 [klein] 4B

  • + Excellent adherence to the specific menu items requested.
  • + Very realistic chalk texture with smudges and natural handwriting style.
  • + Correct date rendering.
  • Several spelling errors in the menu items (e.g., 'Musheram', 'Ootrpous', 'Buter').
  • Small spacing gap in the word 'SPECIALS'.

HiDream I1 Full

  • + Elegant cursive title that reflects the requested 'elegant cursive' style.
  • + Creative addition of a chalk-drawn coffee cup.
  • Completely ignored the requested food items, substituting them with gibberish text.
  • Text appears too uniform and digital, lacking the requested chalk texture variation.
  • The layout is cluttered and the pricing is nonsensical ($2000 for coffee).

Verdict: FLUX.2 [klein] 4B is the clear winner as it attempted and largely succeeded in rendering the specific menu items and price points requested, whereas HiDream I1 Full produced entirely hallucinated gibberish text. Despite some spelling issues in FLUX.2, it captured the authentic texture of a hand-drawn chalkboard much better than the clean, digital look of the HiDream output.

The Reversed Rodeo

Text-to-Image

“Horse riding astronaut in space — horse on top, not vice versa. Surreal, highly detailed, cinematic.”

FLUX.2 [klein] 4B
HiDream I1 Full

AI Judge Analysis

FLUX.2 [klein] 4B

  • + Excellent realism in the horse's anatomy and the astronaut's suit textures.
  • + High level of detail in the background galaxy and the curvature of the Earth.
  • + Clear, cinematic lighting that feels professional.
  • The horse walking on the atmosphere looks a bit stiff compared to a space environment.
  • Lower sense of motion compared to the other model.

HiDream I1 Full

  • + Dynamic sense of movement with the horse appearing to gallop through space.
  • + The white horse provides a nice visual contrast against the dark space background.
  • + The pose of the astronaut holding the reins is more natural.
  • The horse's front left leg has anatomical issues and looks warped.
  • The image has a more 'digital' and less realistic texture compared to its competitor.
  • Details on the space suit are less refined.

Verdict: FLUX.2 [klein] 4B produces a much higher quality image in terms of textures, realism, and background detail, though the pose is somewhat static. HiDream I1 Full captures a more dynamic scene but suffers from significant anatomical distortions in the horse's legs and a lower fidelity finish. FLUX is the preferred choice for its superior clarity and coherent rendering.

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.”

FLUX.2 [klein] 4B
HiDream I1 Full

AI Judge Analysis

FLUX.2 [klein] 4B

  • + Excellent photorealistic lighting and texture on the capybara's fur and jacket.
  • + Strong prompt adherence regarding the businesswoman's facial expression and phone usage.
  • + Authentic interior perspective that feels like professional cinematography.
  • The vehicle's interior dimensions look slightly tight for a standard New York taxi.

HiDream I1 Full

  • + Clear 'TAXI' text on the cap and legible textures.
  • + Good use of the steering wheel to demonstrate the capybara's 'driving' pose.
  • + Vibrant bokeh in the background lights.
  • The businesswoman's face is somewhat blurry and lacks detail.
  • The composition feels slightly more like a collage than a single coherent photo.
  • The capybara's paws look somewhat artificial where they grip the wheel.

Verdict: FLUX.2 [klein] 4B produces a much more realistic and moody image that perfectly captures the requested 'bored' expression of the passenger and the gritty, realistic feel of a night taxi. While HiDream I1 Full is bright and has clear text, it suffers from lower detail in the human subject and a less convincing integration of the animal into the 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.”

FLUX.2 [klein] 4B
HiDream I1 Full

AI Judge Analysis

FLUX.2 [klein] 4B

  • + Excellent atmospheric lighting and moody color palette.
  • + Beautifully detailed thorns and webs border.
  • + Includes all requested elements like twisted trees and cinematic lighting.
  • Significant spelling errors in the Title and location details.
  • Missing a digit in the year (206 instead of 2026).

HiDream I1 Full

  • + The scroll text 'You are invited to a night of frights' is perfectly rendered and legible.
  • + Strong parchment aesthetic and clear composition.
  • + Captures the gothic font style effectively for the main title.
  • Fails to include the specific date, time, and location requested in the prompt.
  • The decorative thorns are less detailed compared to Model A.
  • Includes gibberish text at the bottom instead of the provided event details.

Verdict: Both models struggled with comprehensive text accuracy, but FLUX.2 [klein] 4B is the preferred choice for its superior artistic composition and atmosphere, despite the typos. While HiDream I1 Full managed one line of text perfectly, it completely ignored the specific event details provided in the prompt, whereas FLUX.2 [klein] 4B attempted all of them within a much more immersive and cinematic visual framework.

Bald man challenge

Image Editing
Edit instruction

“Give the person a full, thick head of natural hair with realistic texture, density, and a natural hairline. Preserve facial features and lighting.”

Before After
FLUX.2 [klein] 4B
Before After
HiDream I1 Full

AI Judge Analysis

FLUX.2 [klein] 4B

  • + Excellent preservation of the original facial features and lighting.
  • + The added hair has a realistic texture and matches the style of the person's beard.
  • + Perfectly maintains the background and clothing from the source image.
  • None notable; the edit is highly successful.

HiDream I1 Full

  • + Added some long hair at the back of the head.
  • Completely failed to preserve the original image, changing the person's face, the background, and adding a backpack.
  • Failed to provide a 'full, thick head' of hair, leaving the top bald.
  • Added bizarre artifacts, including a black object fused to the beard/face.

Verdict: FLUX.2 [klein] 4B followed the instructions perfectly, providing a natural and realistic head of hair while keeping the original person and environment intact. HiDream I1 Full failed significantly, generating a different person and background entirely, and did not even add a full head of hair as requested. FLUX.2 is the clear winner for its accuracy and source preservation.

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.”

FLUX.2 [klein] 4B
HiDream I1 Full

AI Judge Analysis

FLUX.2 [klein] 4B

  • + Excellent photorealistic PBR textures on the salmon and rice
  • + Clean typography with 'JAPAN' and 'SUSHI' correctly rendered
  • + Accurate 45-degree isometric perspective
  • The flag icon is incorrect, resembling the flag of Austria or Latvia instead of Japan
  • Missing the requested 'raised diorama base', using a simple plate instead

HiDream I1 Full

  • + Perfect adherence to the 'raised diorama base' and 'isometric miniature' aesthetic
  • + Correct white bold typography for both words
  • + Clean 3D cartoon style with consistent lighting
  • Missing the requested flag icon entirely
  • The textures are very simplified and matte, lacking the 'refined realistic PBR' quality requested

Verdict: FLUX.2 [klein] 4B produces much better textures and text rendering, although it failed to create the 'diorama base' and botched the Japanese flag. HiDream I1 Full followed the composition instructions more closely by including a raised wooden base, but it missed the flag and the textures feel a bit too basic for the prompt requirements. FLUX.2 is the winner for its superior visual clarity and material realism.

Over-the-top cartoon caricature

Editing
Edit instruction

“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.”

Source
FLUX.2 [klein] 4B
HiDream I1 Full

AI Judge Analysis

FLUX.2 [klein] 4B

  • + Successfully captures the 'TV news anchor' setting with a studio background and professional attire.
  • + Includes clear 'dogs' as requested and successfully incorporates 'hockey' visually on the studio monitors in the background.
  • + Strong caricature style with exaggerated features and expressive eyes.
  • The facial resemblance to the source image is slightly reduced by the high level of stylization.
  • Some text in the background is garbled.

HiDream I1 Full

  • + Excellent preservation of the source person's facial features even in a caricature style.
  • + Includes a well-rendered dog character.
  • + Maintains the denim jacket from the original source image.
  • Completely fails to include the 'hockey' request in any meaningful way.
  • The 'TV show anchor' profession is poorly represented, looking more like a living room or office than a studio.
  • Anatomical artifact where the dog merges awkwardly into her collar/neck area.

Verdict: FLUX.2 [klein] 4B is the clear winner because it successfully fulfilled all parts of the multi-faceted prompt, creating a comprehensive scene with a news desk and hockey highlights on the screens. While HiDream I1 Full did a better job at preserving the subject's specific likeness and the clothing of the source image, it failed to incorporate the hockey element and the professional setting 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.”

FLUX.2 [klein] 4B
HiDream I1 Full

AI Judge Analysis

FLUX.2 [klein] 4B

  • + Excellent natural lighting with defined god rays and golden hour sparkles
  • + Characters appear dynamic and in motion as requested by the word 'chasing'
  • + Superior texture detail in the fur and surrounding flowers
  • Missing the baby bunny requested in the prompt
  • Generated two kittens instead of one kitten and one bunny

HiDream I1 Full

  • + Bright, cheerful colors that match the wholesome vibe
  • + Expressive facial features on the animals
  • Failed to include the requested baby bunny, showing two kittens instead
  • Composition is static and lacks the 'chasing' or 'tumbling' action requested
  • Lower image clarity and several anatomical artifacts, such as the fox's ear merging with the background

Verdict: Both models failed to include the full roster of animals, specifically omitting the baby bunny in favor of adding a second kitten. FLUX.2 [klein] 4B is much more successful in capturing the 'hyper-photorealistic' request, providing superior lighting effects and a sense of movement that better aligns with the prompt's action. HiDream I1 Full has a more 'stock photo' appearance with flatter details and less technical polish.

Studio Ghibli Anime Style

Editing
Edit instruction

“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”

Source
FLUX.2 [klein] 4B
HiDream I1 Full

AI Judge Analysis

FLUX.2 [klein] 4B

  • + Excellent preservation of the original poses and expressions.
  • + Soft, watercolor-like pastel palette matches the 'hand-painted' request perfectly.
  • + Accurately replicates the specific patterns and shapes of the original clothing.
  • The facial expressions are softened to the point where the girlfriend's specific look of indignation is lost.

HiDream I1 Full

  • + Strong Studio Ghibli character design aesthetic with expressive eyes and blushing cheeks.
  • + Successfully adds a whimsical, dreamy background consistent with the prompt.
  • Fails to preserve the core 'distracted boyfriend' meme logic; the girlfriend is looking away rather than at the man.
  • Alters the original clothing completely, replacing the plaid shirt with a floral one.

Verdict: FLUX.2 [klein] 4B is the clear winner for image editing as it meticulously preserves the composition, poses, and clothing patterns of the source image while applying a beautiful watercolor Ghibli style. HiDream I1 Full creates a charming illustration, but it fails as an edit by disregarding the specific poses and narrative tension of the original 'distracted boyfriend' meme.

Golden Hour Stroll

Image Editing
Edit instruction

“Add dynamic motion to this photo: make hair blow in the wind, add leaves flying, energetic and lively feel.”

Before After
FLUX.2 [klein] 4B
Before After
HiDream I1 Full

AI Judge Analysis

FLUX.2 [klein] 4B

  • + Excellent preservation of the source image subjects and environment.
  • + Natural placement of flying leaves with realistic shadowing.
  • + Successful modification of the hair to show wind direction while maintaining facial features.
  • The 'lively' feel is subtle compared to Model B.

HiDream I1 Full

  • + High energy and dynamic movement in the pose and environment.
  • + Creative interpretation of 'energetic' by having the model run.
  • Failed to preserve the source image; the dog is missing entirely.
  • The subject's face and clothing have been completely altered.
  • Poor source preservation for an image editing task.

Verdict: FLUX.2 [klein] 4B is the clear winner as it successfully followed the editing instructions while perfectly preserving the original subject, clothing, and background. HiDream I1 Full failed the core requirement of image editing by regenerating a new scene that removed the dog and completely changed the woman's appearance.

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.”

FLUX.2 [klein] 4B
HiDream I1 Full

AI Judge Analysis

FLUX.2 [klein] 4B

  • + Successfully includes the specific name name 'Caffè' and a close attempt at 'Florian'.
  • + Good use of subtle paper texture on the background.
  • + Includes the required banner element with correct dates.
  • Misspells the primary brand name as 'FLAXTION'.
  • Redundant 'Est. 1720' text appearing twice.

HiDream I1 Full

  • + Clean, professional vector aesthetic with high symmetry.
  • + Clever integration of a coffee cup icon inside the cloche.
  • + Excellent typography on the 'Est. 1720' banner.
  • Completely failed to include the primary brand name 'Caffè Florian'.
  • The cloche handle is slightly off-center.

Verdict: Both models followed the stylistic instructions well, but had different failures in prompt adherence. FLUX.2 [klein] 4B attempted the name but failed at spelling it correctly and added redundant text, while HiDream I1 Full produced a cleaner visual but omitted the brand name entirely. HiDream I1 Full is slightly preferred for its superior vector quality and clever iconography, despite the missing text.

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.”

FLUX.2 [klein] 4B
HiDream I1 Full

AI Judge Analysis

FLUX.2 [klein] 4B

  • + Successfully captures a comprehensive flat-vector aesthetic.
  • + Includes more required steps of the mission, specifically including a landing module.
  • + Follows the NASA-inspired color palette very accurately.
  • Text rendering is extremely poor with multiple gibberish words.
  • The rocket design is top-heavy and does not accurately resemble the Saturn V.

HiDream I1 Full

  • + Excellent typography for the main title with a creative logo integration.
  • + High visual impact with strong layout and professional vector lines.
  • + Clean iconography that matches the requested modern style.
  • Failed to include all requested steps, stopping well before the landing phase.
  • Includes literal prompt words like '+ ICON' and '+ UNG' in the text labels.
  • Shows a planet with rings which is out of context for the Apollo 11 moon mission.

Verdict: HiDream I1 Full produces a much more visually polished and professional-looking graphic, but it fails significantly on content, omitting half the requested steps and including hallucinated planetary rings. FLUX.2 [klein] 4B follows the complex mission sequence much better and adheres strictly to the color palette, though its text generation is unreadable. HiDream I1 Full is the likely winner due to its superior artistic execution and the fact that its few legible words are actually correct.

Next steps

Explore each model