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
18.3
arena score
#51 of 62 in Text-to-Image
Skill signature
· Text-to-Image
Stable Diffusion 3.5 Medium
16.8
arena score
#56 of 62 in Text-to-Image
Vote tally
Where the votes landed
HiDream I1 Fast
0%
win rate
Ties
0%
Stable Diffusion 3.5 Medium
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.”
HiDream I1 Fast
Stable Diffusion 3.5 Medium
AI Judge Analysis
HiDream I1 Fast
- + Excellent photorealistic texture and lighting
- + Highly accurate reflections on the glass and sphere
- + Solid geometric composition with clean edges
- − The sphere is quite large, pushing the definition of 'small'
- − The wooden table is somewhat over-smoothed in the foreground
Stable Diffusion 3.5 Medium
- + Successfully rendered a 'small' sphere as requested
- + Good separation of space between the sphere and cube walls
- + Includes a visible plant pot and clear background elements
- − The red book is warped and lacks realistic structural integrity
- − The glass cube rendering is inconsistent, with muddy edges
- − The sphere appears to be levitating unnaturally without a shadow or contact point
Verdict: HiDream I1 Fast produced a significantly more realistic and aesthetically pleasing image with superior lighting and material textures. While Stable Diffusion 3.5 Medium followed the scale of the 'small sphere' better, it failed on the structural details of the book and the clarity of the glass cube.
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.”
HiDream I1 Fast
Stable Diffusion 3.5 Medium
AI Judge Analysis
HiDream I1 Fast
- + Excellent handling of wet pavement reflections
- + High facial detail and natural skin texture
- + Strong cinematic lighting and depth of field
- − Anatomical issue with the man sitting on the bicycle frame rather than the seat
- − Failed to include the requested motion blur on the cars
Stable Diffusion 3.5 Medium
- + More effective use of 'imperfect framing' for a candid feel
- + Better capture of visible rain droplets
- + Good depiction of a worn, realistic red bicycle
- − The man's hands appear mushy and poorly defined
- − Lower resolution/clarity in fine details compared to Model A
- − No motion blur on passing cars as requested
Verdict: Both models failed to incorporate the requested motion blur, but HiDream I1 Fast produced a superior image in terms of clarity and cinematic lighting. While Stable Diffusion 3.5 Medium captured the 'candid' and 'imperfect' atmosphere more effectively, its poor rendering of the subject's hands and face makes it less visually compelling than HiDream I1 Fast.
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.”
HiDream I1 Fast
Stable Diffusion 3.5 Medium
AI Judge Analysis
HiDream I1 Fast
- + Excellent high-contrast lighting with warm torch reflections
- + Highly detailed and legible ornate engraving on the plate armor
- + Clean, sharp facial features and lifelike eye detail
- − The scars look more like face paint rather than physical tissue damage
- − Beads in the hair are a bit generic
- − The hair-armor intersection is a bit messy
Stable Diffusion 3.5 Medium
- + Superb skin texture including dirt, freckles, and more realistic faint scars
- + Excellent portrayal of fire sparks/embers in the background
- + Captures a more intense, 'battle-worn' expression
- − The braids are missing the specific 'small beads' requested in the prompt
- − Armor engraving is slightly less crisp than Model A
- − Lighting on the face is a bit overexposed in certain areas
Verdict: HiDream I1 Fast delivers a more polished, classical fantasy aesthetic with superior armor details and clear beads in the braids. However, Stable Diffusion 3.5 Medium provides a much more lifelike and gritty skin texture that better captures the 'battle-worn' and 'dirt' aspects of the prompt, despite missing the specified 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.”
HiDream I1 Fast
Stable Diffusion 3.5 Medium
AI Judge Analysis
HiDream I1 Fast
- + Features a bold, clear header design with vibrant accents
- + Food photos are consistent in style and high in quality
- + Layout is simple and easy to read at a glance
- − Severely mangled text in the main heading and menu list
- − Failed to include distinct sections for appetizers, pizza, and mains
- − Low utility as multiple photos are almost identical repetitions of pizza
Stable Diffusion 3.5 Medium
- + Excellent adherence to the minimalist grid layout with white background
- + Included diverse food types corresponding to the prompt's sections
- + Better overall composition that resembles a functional, multi-page menu
- − Text is largely illegible gibberish despite better layout
- − Image resolution and sharpness are lower than the competitor
- − Visual clutter makes the individual sections hard to distinguish
Verdict: Stable Diffusion 3.5 Medium followed the layout requirements much more effectively, creating a multi-section grid with diverse food images that fit the modern minimalist prompt. HiDream I1 Fast produced a clearer header and better image quality, but failed the composition by repeating similar pizza images and ignoring the requested menu sub-sections.
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.”
HiDream I1 Fast
Stable Diffusion 3.5 Medium
AI Judge Analysis
HiDream I1 Fast
- + Excellent typography for the main title with a matching fiery glow.
- + Clear starburst containing the correct price.
- + High contrast and vibrant color palette for the burger ingredients.
- − Failed the 'exploded' instruction as the burger remains largely assembled.
- − Text artifacts in the circular graphic on the right.
Stable Diffusion 3.5 Medium
- + Accurate text rendering for all requested phrases.
- + Dramatic and cohesive lighting reflecting off the burger from the flames.
- + Clean, professional ad layout.
- − Failed the 'exploded' instruction by keeping the burger fully stacked.
- − The price is in a radiating line burst rather than a classic starburst shape.
Verdict: Both models failed to deliver the 'exploded' burger view requested in the prompt, opting for a standard hovering burger instead. Stable Diffusion 3.5 Medium is the winner as it correctly followed all text instructions with accurate spelling, whereas HiDream I1 Fast struggled with the secondary text elements.
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.”
HiDream I1 Fast
Stable Diffusion 3.5 Medium
AI Judge Analysis
HiDream I1 Fast
- + Excellent typography legibility for most of the text.
- + Good atmospheric lighting and realistic cafe background.
- + Accurately rendered some of the specific menu items like the Brown Butter Chocolate Chip Cookies.
- − Significant text overlap and clutter in the middle of the menu.
- − Failed to capture the requested elegant cursive calligraphy for the title.
- − Incorrect spelling of 'Butter' and 'Risotto' in various places.
Stable Diffusion 3.5 Medium
- + Beautiful chalkboard texture with realistic smudges and chalk dust.
- + Great artistic flourishes and handwritten style that feels authentic to the medium.
- + Stronger adherence to the 'elegant cursive' requirement for the stylized text.
- − Extreme issues with spelling and word coherence (e.g., 'Trodelu', 'Oocclutes').
- − Failed to include the specific date requested in the prompt.
- − Overcluttered layout with overlapping text boxes on the left.
Verdict: HiDream I1 Fast succeeds in legibility, making most of the menu items readable despite some messy overlapping in the center. Stable Diffusion 3.5 Medium has a much better artistic texture and feels like a real chalk drawing, but its failure to render coherent words or the correct date makes it less useful as a text-driven generation. HiDream I1 Fast is the winner for better prompt adherence regarding specific text strings.
The Reversed Rodeo
Text-to-Image“Horse riding astronaut in space — horse on top, not vice versa. Surreal, highly detailed, cinematic.”
HiDream I1 Fast
Stable Diffusion 3.5 Medium
AI Judge Analysis
HiDream I1 Fast
- + High visual clarity and photographic realism
- + Excellent lighting and realistic textures on the horse's coat
- + Correct character anatomy for both the horse and astronaut
- − Failed the negative constraint regarding position (astronaut is riding the horse)
- − Failed to set the scene in space, placing it in a desert instead
Stable Diffusion 3.5 Medium
- + Successfully followed the setting prompt by placing the subject in space
- + Cinematic composition with Earth in the background
- + Good adherence to the requested color palette and surreal mood
- − Serious anatomical failures with the horse having human-like legs and multiple extra feet
- − Failed the logic check: the astronaut is riding the horse, not vice-versa
- − Visible artifacts and blurring around the horse's legs
Verdict: Both models failed the complex logic instruction to have the 'horse on top' of the astronaut. However, Stable Diffusion 3.5 Medium at least followed the environmental instruction for a space setting, whereas HiDream I1 Fast ignored the space prompt entirely for a desert scene. Stable Diffusion 3.5 Medium is the slight winner for thematic adherence despite its significant anatomical distortions.
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.”
HiDream I1 Fast
Stable Diffusion 3.5 Medium
AI Judge Analysis
HiDream I1 Fast
- + Excellent photographic lighting and crisp city bokeh
- + Realistic textures on the capybara's fur and the woman's coat
- + Successfully captures the bored, distracted expression of the passenger
- − The capybara's head is fused onto a human body with human hands
- − High-angle composition feels less immersive than a dashboard view
Stable Diffusion 3.5 Medium
- + Correctly depicts capybara paws on the steering wheel instead of human hands
- + Dynamic first-person perspective from the dashboard
- + Authentic driver's cap style
- − The passenger is looking forward rather than at her phone
- − The background bokeh is less realistic and distinct than the other model
- − The capybara's facial expression is slightly more startled than professional
Verdict: HiDream I1 Fast produces a more polished social-media-ready aesthetic with superior lighting and passenger details, but it fails significantly by giving the capybara human hands. Stable Diffusion 3.5 Medium adheres better to the physiological requirements of the prompt by using paws, though it misses the specific passenger action. Model B is preferred for accurately following the animal anatomy requested in 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.”
HiDream I1 Fast
Stable Diffusion 3.5 Medium
AI Judge Analysis
HiDream I1 Fast
- + Strong composition with a decorative scroll and centered jack-o-lantern.
- + Includes the requested cobweb borders and twisted trees effectively.
- + Excellent cinematic lighting and spooky atmosphere.
- − Text rendering is messy, especially the lower event details.
- − Major typo in the title 'Halloween Party' and the prompt banner text.
Stable Diffusion 3.5 Medium
- + Artistic border design and clear gothic typography.
- + Good balance of pumpkins and twisted tree silhouettes.
- − Serious spelling errors in the main title ('Halloweeen Invitoween').
- − Failed to render the date correctly (30,1,0..226) and missed the scroll banner element.
- − Failed to centralize the jack-o-lantern as requested.
Verdict: Both models struggled significantly with text legibility and spelling. HiDream I1 Fast is the winner because it followed the compositional cues better, including the central jack-o-lantern and the scroll banner, whereas Stable Diffusion 3.5 Medium failed several layout instructions and produced more nonsensical text strings.
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.”
HiDream I1 Fast
Stable Diffusion 3.5 Medium
AI Judge Analysis
HiDream I1 Fast
- + Excellent adherence to the 3D isometric cartoon style requested.
- + Flawless text rendering of 'JAPAN' and 'SUSHI' with the flag icon.
- + Strong composition with a clear diorama base and nice additional details like chopsticks.
- − The salmon nigiri has a strange cross-section graphic on the side of the rice.
- − The shadows are slightly harsh for a 'gentle lighting' request.
Stable Diffusion 3.5 Medium
- + Features realistic PBR-style textures, especially on the seaweed and roe.
- + Good adherence to the solid light blue background and centered composition.
- − Failed to render the text 'JAPAN' as the primary large bold heading.
- − Missed the 'raised diorama base' and flag icon instructions.
- − Lacks the specific 3D cartoon miniature aesthetic, looking more like a standard render.
Verdict: HiDream I1 Fast followed every part of the complex prompt, including specific text placement, the flag icon, and the miniature diorama base. While Stable Diffusion 3.5 Medium captured the realistic material textures well, it failed to execute the layout and typography requirements correctly.
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.”
HiDream I1 Fast
Stable Diffusion 3.5 Medium
AI Judge Analysis
HiDream I1 Fast
- + Excellent soft lighting and atmospheric god rays.
- + Included two kittens, partially fulfilling the kitten requirement.
- + Superior fur texture and more realistic animal anatomy.
- − Failed to include the baby bunny completely.
- − The animals are sitting still rather than playing and tumbling.
Stable Diffusion 3.5 Medium
- + Successfully included a bunny-like hybrid as requested, though it appears mixed with a cat.
- + Warm and vibrant color palette with clear butterfly inclusions.
- + Good composition with better variety in flower colors.
- − Lacks the tabby kitten requested (all three animals look somewhat canine/feline/fox hybrids).
- − Over-sharpened visual style that looks more like a digital painting than 8K photorealism.
- − The eyes look slightly artificial and glass-like compared to realistic animals.
Verdict: HiDream I1 Fast produces a much more photorealistic image with excellent lighting, but it completely missed the bunny and the 'playful tumbling' action. Stable Diffusion 3.5 Medium tried harder to include the different species but the subjects look like uncanny hybrids of each other, resulting in a less realistic 'masterpiece'. HiDream I1 Fast is the winner for its superior visual quality and technical execution of the fur and lighting.
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.”
HiDream I1 Fast
Stable Diffusion 3.5 Medium
AI Judge Analysis
HiDream I1 Fast
- + Excellent typography rendering for the primary name.
- + Follows the vector emblem style with clean, minimalist lines.
- + Accurately represents the cloche dome shape with stylized steam.
- − Includes a typo in the date, rendering '17210' instead of '1720'.
- − The composition of the steam is a bit floating and detached.
Stable Diffusion 3.5 Medium
- + Highly detailed hand-drawn vintage aesthetic.
- + Pleasant warm monochrome color palette and texture.
- + Creative interpretation of a cloche as the framing element.
- − Poor typography with spelling errors in 'Florrian' and 'Est 170'.
- − Too cluttered to be considered 'minimalist' as requested.
- − Gibberish text on the bottom banner.
Verdict: HiDream I1 Fast is the clear winner for its adherence to the 'minimalist' and 'vector' keywords, producing a clean logo that is much closer to a usable brand asset. While Stable Diffusion 3.5 Medium captures a beautiful vintage texture, it fails significantly on text accuracy and the requested minimalist style.
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.”
HiDream I1 Fast
Stable Diffusion 3.5 Medium
AI Judge Analysis
HiDream I1 Fast
- + Follows the requested layout of defining specific mission steps more clearly.
- + Uses the requested NASA-inspired navy and white palette effectively.
- + The central Saturn V rocket is a strong visual focal point for the infographic theme.
- − Several spelling errors in the text including 'Desceng' and 'Orbit Ring'.
- − The icons for different steps are somewhat repetitive and don't clearly distinguish between Earth and Moon phases.
Stable Diffusion 3.5 Medium
- + Features a more creative and visually interesting vector art style.
- + Captures the 'modern vector' aesthetic with clean lines and a nice use of negative space.
- + Text rendering is stylistically consistent even where words are nonsensical.
- − Completely fails the logical sequence of mission steps, placing 'Landing' as the final icon but with an Earth-like graphic.
- − Includes nonsensical words like 'Trrrapis' and 'Trallity' that don't relate to the prompt.
- − The 'Saturn V' icon requested in phase 1 is missing or replaced by an abstract wireframe.
Verdict: HiDream I1 Fast is the superior choice because it attempts to follow the structured narrative of the Apollo 11 mission steps, whereas Stable Diffusion 3.5 Medium creates a beautiful but logically incoherent layout. HiDream I1 Fast correctly identifies key steps like 'Launch' and 'Landing' in a sequence that makes sense for an infographic, despite some spelling issues.
Next steps
Explore each model
Stability AI's 2.5-billion parameter Multimodal Diffusion Transformer with improvements (MMDiT-X) text-to-image model optimized for consumer hardware, featuring improved image quality, typography, and complex prompt understanding