Head to head
Esc

Models · slot A

to navigate to pick

FLUX.1 [dev] Black Forest Labs GPT Image 1 Mini OpenAI

Settled by community votes across 13 shared challenges, with an AI judge weighing in on each.

FLUX.1 [dev]

24.6 arena score

#16 of 62 in Text-to-Image

Skill signature · Text-to-Image

GPT Image 1 Mini

25.0 arena score

#13 of 62 in Text-to-Image

Vote tally

Where the votes landed

FLUX.1 [dev]

0%

win rate

Ties

0%

GPT Image 1 Mini

0%

win rate

Shared challenges 13

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.1 [dev]
GPT Image 1 Mini

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent photorealism and lighting
  • + Realistic glass thickness and internal reflections
  • + High-quality leather texture on the book
  • The plant is barely visible through the cube due to busy refraction
  • The cube looks more like a solid block than a hollow container

GPT Image 1 Mini

  • + Perfect adherence to all spatial instructions
  • + Clearer visibility of the plant through the glass
  • + Realistic scale of a 'small blue sphere'
  • The lighting is a bit flat compared to Model A
  • Slightly less detailed textures on the book and table

Verdict: Both models followed the prompt accurately, but GPT Image 1 Mini captured the relative scales of the objects more effectively, particularly the small sphere inside a hollow-looking cube. While FLUX.1 [dev] produced a more aesthetically pleasing and photorealistic image with superior lighting, the cube appeared somewhat solid and the sphere occupied most of the internal space.

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.1 [dev]
GPT Image 1 Mini

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent handling of wet pavement reflections and light rain visualization
  • + Strong shallow depth of field effect with realistic bokeh
  • + High fidelity in leather jacket textures and clothing
  • The subject is holding the handlebars rather than repairing the bike
  • The bike looks like a modern road bike rather than a typical utility bike
  • The scene feels slightly more staged than 'candid'

GPT Image 1 Mini

  • + Stronger adherence to the 'repairing' action with hands interactive with the chain
  • + Captures an 'imperfect framing' and close-up perspective that feels very candid
  • + Excellent natural skin texture and facial detail
  • Motion blur on passing cars is minimal and static
  • The bicycle geometry is slightly warped near the rear frame
  • The rain is less visible compared to the other image

Verdict: GPT Image 1 Mini feels much more authentic and candid, successfully capturing the subject in the middle of a repair task with a more naturalistic look. FLUX.1 [dev] produces a technically superior image in terms of environmental lighting and rain effects, but the subject is simply standing next to the bike rather than actively repairing it.

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.1 [dev]
GPT Image 1 Mini

AI Judge Analysis

FLUX.1 [dev]

  • + The skin texture is extremely realistic with visible pores and natural light interaction.
  • + Excellent lighting with a clear distinction between ambient light and highlights.
  • + Sharp focus on the face creates an intimate portrait.
  • The character looks pristine and clean rather than 'battle-worn'.
  • Missing the requested beads in the braids.
  • Armor is barely 'engraved' compared to the other model.

GPT Image 1 Mini

  • + Perfectly captures the 'battle-worn' aesthetic with believable scars and dirt.
  • + Ornate engraving on the armor is highly detailed and fits the prompt exactly.
  • + Lighting and environment feel warmer and more thematic with the 'torchlight' request.
  • The eyes are a bit less lifelike and more matte than model A.
  • The background bokeh is slightly more generic and painterly.

Verdict: GPT Image 1 Mini adhered much better to the specific thematic details of the prompt, particularly the 'battle-worn' look and the intricate armor engravings. While FLUX.1 [dev] produced a stunningly realistic human face, it failed to incorporate the grit, scars, and specific accessories like hair beads that were 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.”

FLUX.1 [dev]
GPT Image 1 Mini

AI Judge Analysis

FLUX.1 [dev]

  • + Elegant, professional layout that looks like a real-world menu
  • + Clear hierarchical organization of menu items and pricing
  • + Features highly appetizing food imagery with creative curved photo integration
  • The text content is mostly gibberish despite the realistic appearance
  • Missing specific requested 'pizza' section category header
  • Layout is two-column rather than a strict 'photos in grid' format

GPT Image 1 Mini

  • + Perfectly adheres to the requested 'grid' for food photos
  • + Includes all three requested section headers: Appetizers, Pizza, and Mains
  • + Clear, bold, and perfectly legible English text
  • The design is overly simplistic and lacks menu item descriptions/prices
  • The layout feels like a placeholder or template rather than a finished professional design
  • The vertical spacing is uneven, leaving large empty voids on the left side

Verdict: FLUX.1 [dev] produced a much more sophisticated and aesthetically pleasing design that captures the 'casual dining' vibe perfectly, though the text is nonsensical. GPT Image 1 Mini adhered more strictly to the prompt's layout constraints like the 2x3 grid and specific section headers, but it failed to include actual menu content, resulting in a very stark and unfinished product. FLUX.1 [dev] is the winner for its superior visual quality and professional composition.

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.”

FLUX.1 [dev]
GPT Image 1 Mini

AI Judge Analysis

FLUX.1 [dev]

  • + Clean photographic detail of the food ingredients
  • + Sophisticated vertical composition
  • Failed to include the prominent 'MAGIC BURGER' title text
  • Missing the requested starburst element for the price
  • Background lacks the requested 'fiery' intensity, opting for small sparkles and ground fire instead

GPT Image 1 Mini

  • + Perfect adherence to all text requirements including 'MAGIC BURGER' title
  • + Successfully integrated the fiery starburst and glowing text effects
  • + Dynamic lighting and background textures more closely match the fiery prompt
  • The burger ingredients are slightly less 'exploded' than Model A, appearing more compact
  • Slightly more digital processing look compared to the clean realism of Model A

Verdict: GPT Image 1 Mini is the clear winner as it followed every instruction in the prompt, including the specific text integration and the fiery starburst which FLUX.1 [dev] completely omitted. While FLUX.1 [dev] produced a high-quality burger render, its failure to include the primary title makes it unsuccessful as an advertisement for 'Magic Burger'.

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.1 [dev]
GPT Image 1 Mini

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent font-like legibility with zero spelling errors.
  • + Includes extra creative details like the 'gluten-free' footer mentioned in passing.
  • + Good background bokeh that creates a café atmosphere.
  • The 'chalk' texture looks more like a digital paint stroke than actual gritty chalk.
  • The handwriting style is a bit too uniform to feel truly $100\%$ hand-drawn.

GPT Image 1 Mini

  • + Outstanding chalk texture that perfectly mimics the dusty, grainy feel of real chalk on a board.
  • + Stronger adherence to the requested handwriting style with natural variations in letter formation.
  • + Correctly rendered all text items requested in the prompt with perfect spelling.
  • The layout for the prices is slightly inconsistent in alignment.
  • The background is less detailed than Model A, showing only a plain wall.

Verdict: While both models followed the prompt perfectly regarding text content and spelling, GPT Image 1 Mini captured the visual texture of chalk and the natural imperfections of handwriting much more realistically than FLUX.1 [dev]. FLUX.1 [dev] produced very clean and legible text, but it felt more like a digital font designed to look like handwriting, whereas GPT Image 1 Mini truly looked like a physical chalkboard.

The Reversed Rodeo

Text-to-Image

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

FLUX.1 [dev]
GPT Image 1 Mini

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent adherence to the specific positioning instruction with the horse riding the astronaut.
  • + Clean, bright cinematic lighting that pops against the background.
  • + Unique and surreal interpretation of the prompt.
  • The white horse's legs have some structural anatomical clipping.
  • Lower background star density makes space feel slightly empty compared to Model B.

GPT Image 1 Mini

  • + High level of skin texture on the horse and fabric detail on the suit.
  • + Rich, atmospheric background with a moon and dense star fields.
  • + Good color grading and cinematic shadows.
  • Failed the core prompt instruction of 'horse on top' (astronaut is riding the horse).
  • Anatomical error with the horse having five legs visible.

Verdict: FLUX.1 [dev] is the clear winner as it successfully followed the difficult 'horse on top' instruction which GPT Image 1 Mini ignored. While Model B has professional-grade textures and lighting, its failure to execute the specific spatial relationship requested and the inclusion of a fifth leg on the horse makes it a lower-quality result for this specific task.

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.1 [dev]
GPT Image 1 Mini

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent photorealism in lighting and textures.
  • + Highly clear and detailed portrayal of both the capybara and the passenger.
  • + The passenger is correctly positioned relative to the driver in the image composition.
  • The passenger appears to be in the front passenger seat rather than the back seat.
  • The capybara's cap looks like a baseball cap instead of a professional driver cap.
  • The capybara's paws are not placed realistically on the steering wheel.

GPT Image 1 Mini

  • + Correct seating arrangement with the passenger clearly in the back seat.
  • + Accurate interpretation of the taxi driver cap and dark jacket.
  • + The capybara's paw placement on the wheel is more anatomically and positionally convincing.
  • Slightly lower image clarity compared to Model A.
  • The passenger is quite dark and blurred, losing much of the facial detail.
  • The framing is slightly more cramped than Model A.

Verdict: While FLUX.1 [dev] produces a sharper image with more vibrant lighting, it fails the spatial requirement of having the passenger in the back seat. GPT Image 1 Mini correctly places the passenger in the rear, captures the professional 'uniform' of the driver better, and maintains the requested bored expression of the human perfectly, making it the more accurate representation of the prompt.

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.1 [dev]
GPT Image 1 Mini

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent thorn border and sharp cinematic lighting.
  • + High contrast and vibrant colors create a modern spooky feel.
  • + Good use of the scroll banner for text element.
  • Several text errors including 'Falloween Rantcl' and 'You Tre'.
  • Extremely repetitive date and time formatting errors.
  • Illustration style leans more towards digital vector than 'vintage parchment'.

GPT Image 1 Mini

  • + Perfect text accuracy for all requested fields.
  • + Captured the 'vintage gothic parchment' texture and mood perfectly.
  • + Includes spider webs as requested in the border detail.
  • Lighting is more muted compared to the 'glowing' request.
  • The 'Date' label was omitted, showing only the numbers.
  • Layout is a bit more standard, though very functional.

Verdict: GPT Image 1 Mini is the clear winner because it successfully rendered all the requested text accurately and captured the specific vintage gothic aesthetic. FLUX.1 [dev] produced a more vibrant illustration, but failed significantly on the text rendering and proofreading, making it unusable as a party invitation.

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.1 [dev]
GPT Image 1 Mini

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent PBR material textures for the sushi fish.
  • + High quality 3D isometric perspective and diorama base.
  • + Beautiful, soft depth-of-field lighting.
  • Failed text rendering with misspelling ('SUSH CATON') and extra characters.
  • Includes unordered garnish not requested in the prompt.

GPT Image 1 Mini

  • + Perfect text rendering of 'JAPAN' and 'SUSHI'.
  • + Accurate 45-degree isometric composition.
  • + Clean, professional graphic design aesthetic.
  • Simple, flat textures compared to the requested realistic PBR materials.
  • The flag icon is stylized as an emoji rather than a standard flag icon.

Verdict: GPT Image 1 Mini followed the layout and text instructions perfectly, delivering clean typography and a professional composition. FLUX.1 [dev] produced significantly better 3D materials and lighting that matched the 'realistic PBR' request, but it failed significantly on the text rendering.

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.1 [dev]
GPT Image 1 Mini

AI Judge Analysis

FLUX.1 [dev]

  • + Features a very clean, illustrative aesthetic with vibrant colors
  • + Excellent lighting effects with clear backlighting on the fur
  • Failed to follow the 'hyper-photorealistic' instruction, producing a 3D-animated/Pixar style instead
  • Missing the kitten requested in the prompt
  • Creatures look more like toys or caricatures than actual animals

GPT Image 1 Mini

  • + Strong adherence to the 'hyper-photorealistic' request with naturalistic fur and textures
  • + Included all four requested animals correctly: puppy, kitten, bunny, and fox kit
  • + Excellent composition showing dynamic movement including leaping and tumbling
  • The fox kit has slightly unusual dark paws that look somewhat muddy
  • The god rays are a bit subtle compared to the request

Verdict: GPT Image 1 Mini captured the prompt's essence perfectly, delivering a hyper-realistic scene with all requested animals in a dynamic, playful arrangement. FLUX.1 [dev] failed significantly on prompt adherence, missing the kitten and opting for a stylized, cartoonish look rather than the requested photorealism.

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.1 [dev]
GPT Image 1 Mini

AI Judge Analysis

FLUX.1 [dev]

  • + Strong minimalist vector emblem aesthetic.
  • + Correct color palette (warm brown/cream) on light background.
  • + Clean, professional-looking illustration of the cloche and steam.
  • Serious spelling errors in the main name ('Flariláan') and secondary text ('Reseaurant').
  • Added extraneous numbers ('11011', '1941') not present in the prompt.

GPT Image 1 Mini

  • + Perfect adherence to text spelling for 'Caffè Florian' and 'Est. 1720'.
  • + Excellent vintage texture and gold-foil effect.
  • + Clear and balanced composition that feels more high-end.
  • Failed the background color instruction, providing a black background instead of light.
  • Typography is a bit bulky for a 'minimalist' request.

Verdict: GPT Image 1 Mini is the preferred choice because it successfully rendered all text correctly, whereas FLUX.1 [dev] produced significant spelling errors. Although GPT Image 1 Mini missed the instruction for a light background, its overall visual quality and text accuracy make it the superior emblem.

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.1 [dev]
GPT Image 1 Mini

AI Judge Analysis

FLUX.1 [dev]

  • + Sophisticated aesthetic using a circular flow that feels like a professional poster.
  • + Includes a dedicated iconography section at the bottom for different mission phases.
  • + Captures the 'NASA-inspired' muted color palette perfectly.
  • Text rendering is mostly gibberish, failing to provide legible descriptions.
  • The sequence is confusing and does not clearly follow the 1-6 logical path requested.

GPT Image 1 Mini

  • + Excellent text legibility and adherence to all six requested steps and descriptions.
  • + Clear, bold vector style that matches the infographic requirement.
  • + Accurate representation of the Lunar Module and Saturn V in a flat style.
  • The 'Translunar' trajectory icon is a messy loop that doesn't make much sense visually.
  • Composition is a bit crowded compared to the more airy Model A.

Verdict: GPT Image 1 Mini is the clear winner because it actually functions as an infographic; it correctly lists all six steps with perfectly legible text and appropriate icons. While FLUX.1 [dev] produced a more visually pleasing and sophisticated art piece, its failure to generate readable text or follow the specific sequence makes it a poor infographic.

Next steps

Explore each model