Reve AI's text-to-image generation model with strong aesthetic quality, accurate text rendering, and detailed instruction following capabilities
Settled by community votes across 4 shared challenges, with an AI judge weighing in on each.
Reve Image 1.0
19.7
arena score
#33 of 44 in Text-to-Image
Skill signature
Not enough comparable category data
The chart appears once both models have ratings across at least three shared arena categories.
Z-Image Turbo
24.7
arena score
#15 of 44 in Text-to-Image
Vote tally
Where the votes landed
Reve Image 1.0
57.1%
win rate
Ties
0.0%
Z-Image Turbo
42.9%
win rate
57.1%
0.0% ties
42.9%
Shared challenges
4
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.”
Reve Image 1.0
Z-Image Turbo
0% wins
0% ties
100% wins
AI Judge Analysis
Reve Image 1.0
- + Excellent photographic quality and sharp focus
- + Realistic caustic lighting effects and reflections on the table
- + The plant is clearly visible through the glass cube as requested
- − The blue sphere is floating/hovering unnaturally inside the cube
- − The lighting direction is somewhat front-on rather than strictly from the left
Z-Image Turbo
- + The blue sphere sits naturally on the bottom of the cube
- + Logical lighting direction from the window on the left
- + Realistic texture on the red book cover
- − The plant is positioned above the cube rather than behind it, making it not visible through the glass
- − The glass cube has some structural inconsistencies at the edges
Verdict: Reve Image 1.0 delivers a much more striking and high-quality image with impressive light refraction on the wooden surface, and it correctly places the plant behind the glass. While Z-Image Turbo handles the physics of the sphere better, it fails the prompt requirement to show the plant through the glass.
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.”
Reve Image 1.0
Z-Image Turbo
0% wins
0% ties
100% wins
AI Judge Analysis
Reve Image 1.0
- + Exceptional engraving detail on the plate armor
- + Subtle and realistic dirt/splatter on the skin
- + Stronger adherence to the 'lifelike eyes' prompt with clear reflections
- − The character looks quite young, which slightly clashes with the 'battle-worn' description
- − Braids look a bit unnatural at the roots
Z-Image Turbo
- + Character model looks appropriately 'battle-worn' and mature
- + Excellent execution of bokeh sparks and torchlight effect
- + Distinctly clear leather straps and cloth underlayer textures
- − Slightly less intricate engraving on the armor compared to Model A
- − The torch flame has some minor digital noise/artifacts at the tip
Verdict: Both models followed the prompt very well, but Z-Image Turbo captures the 'battle-worn' essence more effectively with its character design and the inclusion of visible bokeh sparks. While Reve Image 1.0 has superior armor engraving and skin textures, Z-Image Turbo provides a more cohesive composition with better material variety between the leather, chainmail, and plate.
Bald man challenge
Image EditingEdit instruction
“Give the person a full, thick head of natural hair with realistic texture, density, and a natural hairline. Preserve facial features and lighting.”
Reve Image 1.0
Z-Image Turbo
100% wins
0% ties
0% wins
AI Judge Analysis
Reve Image 1.0
- + Successfully added a full head of hair that matches the beard color
- + Perfectly preserved the background, clothing, and original identity of the subject
- + Maintained the lighting and texture of the original image without hallucinating new elements
- − The hair volume is quite extreme, giving it a slightly wig-like or retro appearance
- − The transition between the hair and the forehead could be softer to look more like a natural hairline
Z-Image Turbo
- + Maintained the subject's identity and facial features well
- − Completely failed to perform the requested edit, leaving the man bald
- − Significantly altered the background landscape and the details of the jacket
- − The glasses frames changed color and style
Verdict: Reve Image 1.0 successfully followed the edit instructions, providing a thick head of hair while keeping the rest of the image identical to the source. Z-Image Turbo failed to add any hair and unnecessarily regenerated the background and clothing, losing the original context of the photo.
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.”
Reve Image 1.0
Z-Image Turbo
100% wins
0% ties
0% wins
AI Judge Analysis
Reve Image 1.0
- + Perfect adherence to the flag request with a correct Japanese flag.
- + Excellent interpretation of 'soft refined textures' with high-quality 3D render look.
- + Clean and bold text rendering with great spacing.
Z-Image Turbo
- + Good isometric composition and clean background.
- + Accurate text rendering for 'JAPAN' and 'SUSHI'.
- + Nice sub-surface scattering effect on the salmon texture.
- − Generated the flag of China instead of the requested Japanese flag.
- − The sushi construction looks slightly illogical with a green element embedded in the rice.
- − Lighting is slightly more washed out compared to the other model.
Verdict: Reve Image 1.0 is the clear winner as it correctly followed the cultural context of the prompt by including the Japanese flag, whereas Z-Image Turbo included the flag of China. Reve Image 1.0 also delivered a more polished 3D aesthetic that better matched the 'refined textures' and 'soft lighting' requested.
Next steps
Explore each model
Tongyi-MAI's 6-billion parameter distilled text-to-image model optimized for speed, achieving high-quality generation in 8 steps or fewer with support for bilingual text rendering