Distilled version of HiDream AI's 17B parameter text-to-image model
Settled by community votes across 13 shared challenges, with an AI judge weighing in on each.
HiDream I1 Fast
#50 of 62 in Text-to-Image
Stable Diffusion 3.5 Large
#28 of 62 in Text-to-Image
Where the votes landed
HiDream I1 Fast
22.2%
win rate
Ties
11.1%
Stable Diffusion 3.5 Large
66.7%
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
HiDream I1 Fast
- + Perfect adherence to spatial instructions with the book on top and sphere inside.
- + Excellent photorealistic textures on the wooden table and red book cover.
- + Clean, minimalist composition with realistic glass reflections.
- − The sphere is quite large, pushing the definition of 'small' in the prompt.
- − The glass cube edges are slightly thin, making it look a bit like an open frame.
Stable Diffusion 3.5 Large
- + Very high detail in the rendering of the book pages and glass thickness.
- + The plant is clearly visible through the glass as requested.
- + Effective use of harsh window light and shadows.
- − Failed the spatial instruction 'On top of the cube sits a red book' by putting the book inside.
- − The blue sphere is floating unnaturally above the book.
Verdict: HiDream I1 Fast followed all spatial instructions perfectly, placing the sphere inside the cube and the red book on top. Stable Diffusion 3.5 Large failed the primary layout request by placing both objects inside the cube, though it featured higher detail in textures and lighting.
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
HiDream I1 Fast
- + Excellent shallow depth of field effect
- + Beautiful and realistic reflections on the wet pavement
- + High facial detail and natural skin texture
- − The man appears to be sitting on or hovering over the bike rather than repairing it
- − Missing the requested motion blur on the passing cars
- − The bike structure is slightly warped at the handlebars
Stable Diffusion 3.5 Large
- + More active posture that better matches the 'repairing' instruction
- + Realistic depiction of falling rain drops
- + Includes better street level atmosphere and accurate bicycle proportions
- − The white car in the background lacks the requested motion blur
- − The arm skin texture is slightly over-sharpened and unnatural in light
- − The framing is very centered for a prompt requesting 'imperfect framing'
Verdict: HiDream I1 Fast produces a more cinematic portrait with superior depth of field and beautiful lighting, but the subject isn't clearly 'repairing' the bike. Stable Diffusion 3.5 Large follows the action of the prompt better, showing a man bent over his work, and captures a more gritty, realistic rainy day atmosphere despite some harsh textures on the subject's skin.
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
HiDream I1 Fast
- + Excellent depiction of the requested beads in the braids
- + Strong use of warm torchlight and bokeh sparks
- + Intricately detailed engraving on the plate armor
- − The facial scars look painted on rather than realistic skin texture
- − The lighting on the face is a bit oversaturated and flat
Stable Diffusion 3.5 Large
- + Extremely realistic skin texture with lifelike dirt and scars
- + Superior rendering of the cloth and chainmail underlayers
- + More natural and cinematic lighting
- − Missed the request for beads in the hair braids
- − The background characters distract slightly from the 'close portrait' feel
Verdict: HiDream I1 Fast followed the specific detail of the beads in the hair much better and captured the fantasy 'paladin' aesthetic with vibrant torchlight. However, Stable Diffusion 3.5 Large produced a significantly more lifelike image with superior textures on the skin, armor, and fabric, making it the more visually impressive result despite missing the beads.
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
HiDream I1 Fast
- + Strong bold header with a clear 'Restaurant Menu' title.
- + Good color contrast and vibrant red accent bar.
- + Successfully includes pizza images as requested.
- − Very poor text rendering for the menu items.
- − Limited food variety, mostly repeating the same pizza image.
- − Fails to create distinct sections for appetizers, pizza, and mains.
Stable Diffusion 3.5 Large
- + Excellent minimalist aesthetic with professional typography.
- + Sophisticated grid layout that integrates various food photos effectively.
- + Follows the prompt instructions for specific sections better than the competitor.
- − The food photos are predominantly pizza despite section headers for mains and appetizers.
- − Small text becomes illegible and contains gibberish characters.
Verdict: Stable Diffusion 3.5 Large is the clear winner as it produces a professional, modern layout that looks like a real minimalist menu. While HiDream I1 Fast has a bold header, the body of the menu is poorly structured and the text is unreadable, whereas Stable Diffusion 3.5 Large excels in composition and design elegance.
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
HiDream I1 Fast
- + Excellent text rendering and placement that follows all requirements
- + Vibrant colors and high photorealistic detail on the ingredients
- + Accurately represents and places the 'starburst' price tag
- − The burger is not truly 'exploded' with all components suspended; it is mostly stacked
- − The background is slightly less dynamic than requested
Stable Diffusion 3.5 Large
- + Strong sense of heat and fire with great interaction between the burger and embers
- + High visual quality and realistic textures on the patty and cheese
- − Failed to include any of the requested text
- − Missing the starburst graphic for the price entirely
- − The burger is stacked rather than exploded as chips or components in mid-air
Verdict: HiDream I1 Fast is the clear winner because it successfully integrated all requested text and graphic elements, including the complex starburst and price tag, with high legibility. Stable Diffusion 3.5 Large produced a high-quality image of a burger on fire, but completely ignored the text and advertisement layout requirements of the prompt.
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
HiDream I1 Fast
- + Excellent chalk texture and realistic handwritten variations.
- + Close adherence to the specific date requested.
- + Correct pricing and item titles for the first several menu items.
- − Significant text overlap and jumbled characters in the middle section.
- − Contains a spelling error with 'Buter' instead of 'Butter'.
Stable Diffusion 3.5 Large
- + Provides a highly realistic and aesthetic cafe setting/atmosphere.
- + Good layout and composition for a commercial environment.
- + Clean and legible title font.
- − Failed the date request, showing 2024 instead of 2026.
- − The menu text is largely gibberish with many spelling errors like 'Todaay' and 'Cholcalte'.
- − Font style looks more digital/printed than hand-drawn chalk.
Verdict: HiDream I1 Fast better captures the specific handwriting style and chalk texture requested in the prompt, despite some messy overlapping in the center of the board. Stable Diffusion 3.5 Large creates a beautiful environmental shot, but fails significantly on text accuracy, date adherence, and 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
HiDream I1 Fast
- + High resolution and clean rendering
- + Good anatomical correctness of the horse and astronaut
- − Completely failed the environment prompt (desert instead of space)
- − Failed the requested inversion (astronaut is on top, despite prompt asking for horse on top)
- − Lacks the requested surreal qualities
Stable Diffusion 3.5 Large
- + Successfully captured the space environment and cinematic lighting
- + Highly detailed cosmic elements and clouds
- + Expressive and dynamic composition
- − Failed the requested inversion (astronaut is on top, despite prompt asking for horse on top)
- − Muddled anatomy where the horse legs meet the cosmic clouds
Verdict: Both models failed the specific spatial reasoning instruction to put the horse on top of the astronaut, likely due to strong training bias for the common 'astronaut on horse' trope. However, Stable Diffusion 3.5 Large is the superior image as it correctly placed the subjects in space with cinematic detail, whereas HiDream I1 Fast ignored the environment prompt and generated a standard desert scene.
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
HiDream I1 Fast
- + Successfully includes the businesswoman in the back seat as requested
- + Rendered the capybara head and texture realistically
- + The urban background lighting is cinematic and fits the New York theme
- − The driver has human hands instead of capybara paws
- − The capybara is positioned in the wrong seat for a left-hand drive vehicle or lacks a clear dashboard
Stable Diffusion 3.5 Large
- + Correctly depicted capybara paws on the steering wheel
- + Very detailed clothing textures and expressive animal facial rendering
- + Strong vibrant colors and crisp interior car details
- − Completely failed to include the passenger in the back seat
- − The capybara appears to have human legs in jeans, which is anatomically confusing
Verdict: HiDream I1 Fast followed the complex prompt instruction to include the bored businesswoman in the backseat, whereas Stable Diffusion 3.5 Large completely ignored this part of the scene. Although Stable Diffusion 3.5 Large correctly rendered the capybara's paws on the wheel, HiDream I1 Fast is the superior image for capturing the narrative and humor of the requested 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
HiDream I1 Fast
- + Successfully included all event details (date, time, location).
- + Excellent layout with a clearly legible gothic title.
- + Strong adherence to the central jack-o-lantern requirement.
- − Text on the middle scroll banner and bottom details contains minor artifacts and typos.
- − The background trees and webs are somewhat simplistic in style.
Stable Diffusion 3.5 Large
- + Very atmospheric and high-quality vintage paper texture.
- + Beautiful cinematic lighting with a glowing moon and detailed trees.
- + The scroll banner text is quite legible and decorative.
- − Failed to include the specific event details (Date, Time, Location) at the bottom.
- − The jack-o-lantern is small and off-center rather than central.
- − The overall image is not a square format as requested.
Verdict: HiDream I1 Fast followed the complex instructions much more accurately, successfully including all the specific text details for the invitation, despite some minor character artifacts. Stable Diffusion 3.5 Large produced a more visually striking and atmospheric image, but failed significantly on the specific text requirements and the requested square aspect ratio.
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
HiDream I1 Fast
- + Excellent adherence to the 'top-center large bold text' layout requirement.
- + Very clean, minimal 3D cartoon aesthetic with soft refined textures.
- + Accurate isometric perspective and solid blue background.
- − The sushi roll inside the nigiri is a nonsensical culinary depiction.
- − Included extra chopsticks that were not specifically requested.
Stable Diffusion 3.5 Large
- + Highly realistic PBR materials, especially on the salmon and rice grains.
- + Greater variety of sushi types within the miniature scene.
- + Captures the 'diorama' feel well with the small flag.
- − Failed the text placement instructions, putting text on a card rather than 'top-center' large and bold.
- − The layout is more cluttered than the requested 'minimal garnish' design.
- − Includes a strange bowl of pink salt/rice in the background.
Verdict: HiDream I1 Fast followed the complex layout instructions much more closely, placing the text exactly where requested with a clean graphic design. While Stable Diffusion 3.5 Large produced more realistic materials and textures, it ignored the specific text placement and sizing constraints, resulting in a cluttered composition compared to HiDream's minimalist isometric look.
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
HiDream I1 Fast
- + Excellent fur texture rendering and soft lighting.
- + Expressive facial features on the puppy and fox.
- + Warm, cohesive color palette.
- − Missing the baby bunny entirely.
- − Animals are sitting still instead of 'playfully chasing and tumbling'.
- − The proportions of the kittens relative to the puppy seem inconsistent.
Stable Diffusion 3.5 Large
- + Successfully included all four requested animals: puppy, kitten, bunny, and fox.
- + Dynamic composition that captures the 'playfully chasing' action requested.
- + Beautiful implementation of dew sparkles and god rays.
- − The fox's facial features are slightly blurry and less detailed than other animals.
- − Artifacts on the bottom-most butterfly (looks like a smear).
Verdict: Stable Diffusion 3.5 Large is the clear winner as it followed all prompt instructions, including the presence of the bunny and the active 'chasing' motion. HiDream I1 Fast produced a high-quality static portrait but failed to include the bunny or the requested sense of movement.
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
HiDream I1 Fast
- + Clean vector illustration style
- + Correct spelling of 'Caffè'
- + Strong color contrast and composition
- − Typos in the numerical date ('17210')
- − The steam elements are a bit disjointed and messy
Stable Diffusion 3.5 Large
- + Excellent texture and vintage aesthetic
- + Correct numerical date '1720'
- + Minimalist and balanced layout
- − Spelling error in the main text ('Cafféé')
- − Stylized steam looks a bit cartoonish compared to the vintage theme
Verdict: HiDream I1 Fast delivers a cleaner vector emblem with better central typography, but fails on the historical date requested. Stable Diffusion 3.5 Large has a superior vintage feel and correct date, but fails on the spelling of the restaurant name. HiDream I1 Fast is slightly better for professional use as the date is an easier fix than primary branding 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.”
AI Judge Analysis
HiDream I1 Fast
- + Attempts to follow the specific 6-step chronological structure
- + Large, readable main title
- + Includes relevant iconography like the American flag and lunar module
- − Numerous spelling errors in text labels
- − Graphic style is messy with jagged lines rather than clean vectors
- − The central rocket asset has floating, disjointed pieces
Stable Diffusion 3.5 Large
- + Excellent adherence to the requested NASA-inspired color palette
- + Professional, balanced composition with a clean vector aesthetic
- + High level of detail and fine lines suitable for a poster
- − Fails to follow the specific 6-step chronological list requested
- − The main rocket depicted is a generic shuttle/spaceplane rather than the Saturn V
- − Text consists almost entirely of illegible gibberish
Verdict: HiDream I1 Fast better followed the logical structure of the prompt by attempting to include the named mission steps, but it suffered from poor image quality and severe spelling errors. Stable Diffusion 3.5 Large produced a much more visually stunning and stylistically accurate infographic poster, but it ignored the specific list of steps and the request for a Saturn V rocket. Stable Diffusion 3.5 Large is the preferred choice for its superior professional aesthetic and color adherence.
Explore each model
Stability AI's 8.1-billion parameter Multimodal Diffusion Transformer (MMDiT) text-to-image model featuring improved image quality, typography, complex prompt understanding, and resource-efficiency