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
OmniGen v2
16.8
arena score
#57 of 62 in Text-to-Image
Vote tally
Where the votes landed
HiDream I1 Fast
0%
win rate
Ties
0%
OmniGen v2
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
OmniGen v2
AI Judge Analysis
HiDream I1 Fast
- + Excellent realism in the glass textures and reflections.
- + Accurate representation of the requested lighting from the left.
- + Clean, modern composition with high clarity.
- − The sphere is quite large, pushing the definition of 'small' relative to the cube.
- − The sphere appears to be levitating slightly rather than resting on the bottom surface.
OmniGen v2
- + Perfectly adheres to the scale requested for a 'small' sphere.
- + High levels of photographic realism in the book texture and table grain.
- + Excellent spatial arrangement and depth.
- − The plant is positioned more to the side than directly behind the cube as visible through the glass.
- − The perspective on the cube's top edge is slightly distorted under the book.
Verdict: Both models followed the prompt instructions very well, including the spatial requirements. OmniGen v2 is the winner because it adhered more accurately to the scale of the 'small' sphere and provided a more natural placement of objects, whereas HiDream I1 Fast made the sphere much larger than requested.
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
OmniGen v2
AI Judge Analysis
HiDream I1 Fast
- + Excellent adherence to the 'cinematic but realistic' and 'candid' atmosphere
- + Great execution of reflections on wet pavement and shallow depth of field
- + Includes passing cars with light trails/blur as requested
- − Physical interaction with the bike is slightly awkward with hand placement
- − Bicycle geometry has some minor structural inconsistencies near the handlebars
OmniGen v2
- + Natural skin textures on the man's face
- + Good realization of the red bicycle and light rain streaks
- + Clear and sharp main subject
- − Missed the 'motion blur from passing cars' requirement
- − The bicycle design is inconsistent, with a missing chain/pedal connection to the rear wheel
- − Composition feels less 'candid' and more like a staged portrait
Verdict: HiDream I1 Fast better captures the specific technical requirements of the prompt, particularly the 50mm lens look, motion blur from cars, and the moody cinematic atmosphere. While OmniGen v2 has good skin texture, it fails to include the requested motion blur and produces a bicycle with significant mechanical errors.
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
OmniGen v2
AI Judge Analysis
HiDream I1 Fast
- + Excellent depiction of ornate engraved plate armor with high detail
- + Strong adherence to the 'battle-worn' aesthetic with realistic lighting reflections
- + Rich texture on hair braids and beads
- − The scars on the forehead look like red paint or fresh cuts rather than faint scars
- − Shadows on the face are a bit harsh in some areas
OmniGen v2
- + Soft, realistic skin texture and lifelike eyes
- + Clean bokeh effects and good shallow depth of field
- + Accurate representation of the hair beads
- − The 'dirt' on the face appears more like clean freckles or beauty marks than battle grime
- − The armor engraving is much simpler and less 'ornate' than requested
- − Lacks the 'battle-worn' intensity described in the prompt
Verdict: HiDream I1 Fast captures the gritty, battle-worn essence of a paladin much more effectively, particularly with the highly detailed engravings on the armor and the presence of dirt and wear. OmniGen v2 produces a cleaner, more aesthetically 'pretty' image that ignores the battle-worn requirement, making the character look too pristine for the context.
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
OmniGen v2
AI Judge Analysis
HiDream I1 Fast
- + Matches the grid layout for food photos effectively.
- + Follows the request for a bold header section.
- + The food images are relatively consistent in style.
- − Significant text corruption and spelling errors even in main headers.
- − The layout feels cramped and slightly dated rather than modern minimalist.
- − Menu items and prices are illegible.
OmniGen v2
- + Successfully creates a modern, clean, and airy minimalist aesthetic.
- + Features vibrant color blocks and high-quality product photography.
- + Layout is more professional and balanced with distinct sections.
- − Contains spelling errors in headers (e.g., 'RESTAURATED MENTS').
- − Text in the body of the menu is represented by placeholder-like squiggles.
Verdict: OmniGen v2 is the clear winner for its superior grasp of 'modern minimalist' design principles, providing a much cleaner and more professional layout. While both models struggled with text accuracy, OmniGen v2 produced much higher quality food photography and a more appealing color palette compared to HiDream I1 Fast's cluttered and distorted output.
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
OmniGen v2
AI Judge Analysis
HiDream I1 Fast
- + Excellent text integration for the price starburst and main title
- + Includes all requested ingredients with high photorealistic detail
- + Sophisticated use of lighting and embers to create an atmospheric scene
- − The burger is not truly 'exploded' as requested, appearing mostly stacked
- − The secondary 'Limited Time' text circle is slightly garbled and duplicate
OmniGen v2
- + Bold and clear graphic design for the logo and layout
- + Accurate text rendering for the main titles and price
- + Strong fiery glow effect that matches the prompt's theme
- − Failed the 'exploded' instructions as the burger is a standard double stack
- − The visual style is more illustrative/cartoonish than photorealistic
- − Missing several ingredients like onions and sauce visible in Model A
Verdict: HiDream I1 Fast delivers a much more photorealistic and appetizing image that fits the 'advertising' aesthetic well, despite failing to truly 'explode' the burger components. OmniGen v2 has cleaner text but takes a very simplified, illustrative approach that ignores several requested ingredients. HiDream I1 Fast is the winner for its superior texture detail and atmospheric composition.
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
OmniGen v2
AI Judge Analysis
HiDream I1 Fast
- + Excellent environmental context with a cozy café setting and natural lighting.
- + Sophisticated cursive title that captures the 'elegant' prompt requirement.
- + Realistic chalk texture including dust smudges and varying stroke opacity.
- − Significant text artifacts where numbers and words overlap mid-board.
- − Misspelling of 'Butter' as 'Buter'.
OmniGen v2
- + High legibility of individual chalk letters throughout the entire board.
- + Includes additional thematic text and more balanced line spacing.
- + Accurately follows the date requirement and most of the menu items.
- − Flat, studio-style composition lacks the 'cozy café' atmosphere requested.
- − Numerous spelling errors such as 'Musonhom', 'Lemont', and 'glite free'.
- − Handwriting style is quite blocky and lacks the 'elegant cursive' requested for the title.
Verdict: HiDream I1 Fast produces a much more visually pleasing and realistic photograph with convincing environmental details, but it struggles with severe text overlapping and jumbled characters in the center of the board. OmniGen v2 provides a much cleaner layout for reading, though it lacks the requested café background and suffers from multiple spelling errors. HiDream I1 Fast is the likely winner because it captures the requested aesthetic and 'elegant cursive' style, whereas OmniGen v2 feels like a flat graphic with several typos.
The Reversed Rodeo
Text-to-Image“Horse riding astronaut in space — horse on top, not vice versa. Surreal, highly detailed, cinematic.”
HiDream I1 Fast
OmniGen v2
AI Judge Analysis
HiDream I1 Fast
- + High level of realism and photographic texture in the horse and environment.
- + Natural-looking dust and lighting effects consistent with a cinematic style.
- + Higher resolution feel and anatomically plausible horse.
- − Failed completely on the environment prompt, placing the subjects in a desert instead of space.
- − Failed to follow the specific spatial instruction (horse on top was interpreted as riding normally).
OmniGen v2
- + Successfully followed the 'space' setting with stars and celestial bodies.
- + Good clarity on the astronaut suit details and clean colors.
- − Failed the specific spatial instruction 'horse on top, not vice versa'.
- − The anatomy of the back legs is warped and lacks proper connection.
- − Styling is more illustrative/digital than 'cinematic' or 'highly detailed'.
Verdict: Both models failed the specific prompt request for a 'horse on top' of the astronaut, providing a standard riding scene instead. HiDream I1 Fast produced a much more realistic and cinematic image, but ignored the 'space' setting entirely, whereas OmniGen v2 correctly placed the scene in space but suffered from significant anatomical defects in the horse's legs.
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
OmniGen v2
AI Judge Analysis
HiDream I1 Fast
- + Successfully places the businesswoman in the back seat as requested
- + Rendered a 'TAXI' label on the cap clearly
- + High level of detail in the capybara's fur and the city bokeh lights
- − The capybara has human hands instead of capybara paws
- − The passenger and driver appear to be sitting side-by-side due to perspective issues
OmniGen v2
- + Clean, vibrant colors and sharp focus on the subjects
- + Good execution of the jacket and hat on the capybara
- − Failed the spatial prompt by putting the businesswoman in the front passenger seat
- − The capybara has human hands gripping the wheel
- − The steering wheel appears to be on the passenger side or centrally misaligned
Verdict: HiDream I1 Fast followed the prompt more accurately by attempting to place the woman in the back seat, even though the perspective is slightly flat. OmniGen v2 failed the positioning instruction by placing the woman in the front seat and also suffered from significant anatomical and structural logic errors regarding the steering wheel and hands. HiDream I1 Fast is the winner for better prompt adherence and environmental rendering.
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
OmniGen v2
AI Judge Analysis
HiDream I1 Fast
- + Strong composition with a central focus on the pumpkin
- + Better execution of the parchment paper texture and weathered edges
- + Most date and location information is legible.
- − Banner text is garbled and incomplete
- − Spelling errors in the location name at the bottom
OmniGen v2
- + Beautiful title typography and polished lighting
- + Includes all requested elements like thorns and glowing moon background
- + Clever use of negative space for the pumpkin carving
- − Banner text is completely illegible
- − Information at the bottom is scrambled and repetitive
- − Fails to place requested text onto the scroll banner correctly
Verdict: HiDream I1 Fast delivers a more traditional invitation layout that is easier to read, though it struggles with the banner text. OmniGen v2 has superior visual polish and cinematic lighting but fails significantly on text legibility and following the specific placements for the event details. HiDream I1 Fast is the likely winner for better adherence to the informational requirements of the 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.”
HiDream I1 Fast
OmniGen v2
AI Judge Analysis
HiDream I1 Fast
- + Excellent text rendering with clean, professional typography.
- + Higher quality 3D materials that feel like a physical diorama.
- + Includes more relevant props like chopsticks and a realistic Japanese flag icon.
- − The sushi design is slightly anatomically strange, with a piece of filling visible inside the nigiri rice.
- − The composition is a bit tight on the vertical axis.
OmniGen v2
- + Perfectly follows the 45-degree isometric camera angle.
- + Very vibrant, clean cartoon aesthetic that perfectly matches the 'miniature' request.
- + Very clean, centralized composition.
- − Failed to render a correct Japanese flag, providing a generic blue/yellow/red rectangle instead.
- − The text is slightly less polished in its integration with the background.
Verdict: HiDream I1 Fast produced a superior image with highly refined textures and perfectly rendered text, though it took a slight creative liberty with the sushi structure. OmniGen v2 nailed the isometric perspective and simple cartoon lighting, but failed significantly on the flag icon. HiDream I1 Fast is the winner for its professional material quality and better adherence to the specific text and icon requirements.
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
OmniGen v2
AI Judge Analysis
HiDream I1 Fast
- + Excellent fur texture rendering and realistic color palettes
- + Captures the 'gold rays' and atmosphere accurately
- + Good anatomical details on the kitten and fox kit
- − Failed to include the baby bunny
- − Replaced the bunny with a second kitten
OmniGen v2
- + Clearer, more saturated colors for the wildflowers
- + Includes distinct animal characters with very expressive features
- − Has a very 'AI-cartoon' or 3D-render look rather than the requested hyper-photorealistic style
- − Missing several animals from the prompt (fox kit and baby bunny are missing/blashed together into one hybrid creature)
- − Anatomical inconsistencies in the limbs and faces
Verdict: HiDream I1 Fast is the clear winner as it adheres much closer to the 'hyper-photorealistic' style requested, whereas OmniGen v2 produced a stylized, cartoonish image. While HiDream I1 Fast accidentally included two kittens instead of a bunny, the overall visual quality, lighting, and fur details are significantly superior to OmniGen v2.
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
OmniGen v2
AI Judge Analysis
HiDream I1 Fast
- + Successfully included all prompt elements including the banner and steam.
- + Accurate spelling of 'Caffè Florian' with the correct accent mark.
- + Professional vector aesthetic with good use of golden-brown color gradients.
- − Typo in the date banner, showing '17210' instead of '1720'.
- − The steam effect is a bit cluttered and overlaps with the framing circle.
OmniGen v2
- + Clean minimalist vector style that adheres well to the 'vintage logo' request.
- + Excellent legibility and font choice for the year banner.
- + Centralized and balanced composition.
- − Failed to spell the restaurant name correctly, outputting 'CAFFFLORIN'.
- − The cloche design is very basic and lacks the detail seen in the competitor.
Verdict: HiDream I1 Fast delivers a much more sophisticated illustration and handles the primary brand name perfectly, though it stumbles on the numeric date. OmniGen v2 has a cleaner minimalist layout but fails significantly on the text rendering of the restaurant name. HiDream I1 Fast is the preferred choice for its superior artistic detail and accurate brand spelling.
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
OmniGen v2
AI Judge Analysis
HiDream I1 Fast
- + Follows the layout instructions by organizing the infographic as steps around a central image.
- + Correctly includes the requested color palette and features a recognizable Saturn V rocket launch.
- + The text 'APOLLO 11' is clearly and correctly rendered at the top.
- − Significant spelling errors in the labels for each step (e.g., 'DESCENG', 'EACH OPBIT').
- − Iconography is somewhat messy and inconsistent with the flat-vector style requested.
- − Fails to cleanly represent all 6 specific requested steps in the correct order.
OmniGen v2
- + Much cleaner aesthetic with a professional flat-vector design that adheres well to a modern UI style.
- + Excellent adherence to the color palette and very crisp, consistent iconography.
- + Text rendering is sharper, even if the content is incorrect.
- − Failed basic prompt instructions by titling the poster 'APOLO 17' instead of 'APOLLO 11'.
- − The labels for the steps are largely nonsensical gibberish ('LANDEN ATE COLINS', 'TRANOIMON').
- − Labels the organization as 'NSA' instead of using the NASA-inspired theme correctly.
Verdict: HiDream I1 Fast followed the structural layout of an infographic much better and correctly titled the mission, though its icons were cluttered and its spelling was poor. OmniGen v2 achieved a far superior professional vector aesthetic, but failed the most basic prompt requirements by misidentifying the mission as Apollo 17 and filling the poster with gibberish. HiDream I1 Fast is the winner for actually representing the correct historical mission and following the instructional steps, despite the lower graphical quality.
Next steps
Explore each model
Unified multimodal model for text-to-image generation, instruction-guided image editing, personalized generation, and virtual try-on