Black Forest Labs' precision image generation model with maximum control, reliable text rendering, and complete creative control supporting up to 4MP output
Settled by community votes across 8 shared challenges, with an AI judge weighing in on each.
FLUX.2 [flex]
#14 of 62 in Text-to-Image
Reve Image 1.0
#42 of 62 in Text-to-Image
Where the votes landed
FLUX.2 [flex]
100.0%
win rate
Ties
0.0%
Reve Image 1.0
0.0%
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
FLUX.2 [flex]
- + Perfect adherence to spatial instructions with the plant behind the cube.
- + High level of photo-realism and clean glass rendering.
- + Accurate and aesthetically pleasing soft window lighting from the left.
- − The blue sphere is relatively large compared to the requested 'small' sphere.
Reve Image 1.0
- + Good photographic texture on the wooden table and red book.
- + The sphere is floating, which creates an interesting visual.
- − Failed the spatial 'behind' prompt as the plant appears to be to the left/beside the cube rather than visible through it.
- − The glass cube has dark teal edges that look somewhat artificial compared to clear glass.
- − The sphere is floating without explanation, which may not have been the intent.
Verdict: FLUX.2 [flex] followed the prompt instructions much more accurately, specifically placing the plant behind the cube so that it is visible through the glass. Reve Image 1.0 struggled with the spatial relationship of the objects and the transparency of the glass, placing the plant mostly to the side and creating a floating effect for the sphere that wasn't requested.
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
FLUX.2 [flex]
- + Excellent execution of all prompt elements including beaded braids and dirt textures
- + Highly detailed engraving on the plate armor with realistic metal reflections
- + Superior skin texture and lifelike eyes
- − The scars appear a bit like superficial scratches rather than deep battle scars
Reve Image 1.0
- + Strong cinematic lighting and atmospheric bokeh
- + Intricate engraving on the armor
- + Good representation of the braided hair and beads
- − The subject looks very young, which slightly clashes with the 'battle-worn' descriptor
- − The dirt on the face lacks depth and looks a bit like digital noise/artifacts
Verdict: FLUX.2 [flex] wins because it captures the 'battle-worn' essence much more effectively with a more mature character and significantly better texture work on the skin and leather. While Reve Image 1.0 has nice lighting and intricate armor, FLUX's attention to detail across the clothing, braided beads, and lifelike eyes makes it the superior output.
Bald man challenge
Image Editing“Give the person a full, thick head of natural hair with realistic texture, density, and a natural hairline. Preserve facial features and lighting.”
AI Judge Analysis
FLUX.2 [flex]
- + Excellent preservation of the original facial features and expression.
- + Seamless blending of the new hair with the existing sideburns and beard.
- + Realistic, short texture that fits the rugged aesthetic of the image.
- − The hairline transition on the forehead is slightly harsh in some spots.
Reve Image 1.0
- + Successfully added a very 'full and thick' head of hair as requested.
- + Matches the hair color well to the existing beard.
- − The hair volume looks slightly unnatural and wig-like.
- − Subtly alters the shape of the face/forehead compared to the source.
Verdict: FLUX.2 [flex] produced a much more convincing and realistic result by adding hair that perfectly matched the lighting and texture of the original subject's beard. While Reve Image 1.0 followed the prompt for a 'thick' head of hair, the resulting style felt less natural and slightly modified the subject's facial structure.
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
FLUX.2 [flex]
- + Perfect text placement and rendering according to the prompt.
- + Excellent miniature aesthetic with realistic clay-like PBR textures.
- + Comprehensive sushi variety including nigiri and rolls.
- − The flag icon is slightly large relative to the text.
- − Lighting is a bit flat compared to the soft shadows requested.
Reve Image 1.0
- + Very clean, minimalist 3D cartoon style.
- + Nice glossy material on the plate and fish tops.
- + Good centered composition.
- − Text layout does not follow the requested 'SUSHI below it' instruction strictly (placed flag on same line).
- − The green garnish is oversized and awkwardly clipping into the sushi.
- − Missing maki rolls or additional variety typical of a Japan sushi theme.
Verdict: FLUX.2 [flex] followed the prompt instructions precisely, particularly with the text layout and the variety of sushi pieces. It captured the diorama aesthetic more effectively, whereas Reve Image 1.0 struggled with the text arrangement and had a less polished interaction between the sushi model and its garnishes.
Over-the-top cartoon caricature
Editing“Create a caricature of me and my job. Make it exaggerated and humorous, incorporating my profession as a tv show anchor and my love for dogs and hockey.”
AI Judge Analysis
FLUX.2 [flex]
- + Excellent inclusion of multiple dogs in hockey jerseys, satisfying the prompt well.
- + The character design is a classic caricature style with a large head and small body.
- + Maintains the subject's outfit (denim shirt) from the source image.
- − The facial likeness to the source image is lower than the competitor, appearing more generic.
- − Small text on background screens is garbled.
Reve Image 1.0
- + Exceptional facial likeness to the source image while maintaining a caricature style.
- + Cleverly integrates hockey and news anchor professions through the jersey/blazer combo.
- + High-quality rendering with vibrant colors and clear central text.
- − The hockey stick is floating/disconnected from the desk base.
- − Includes only one dog despite the prompt mentioning a love for 'dogs' (plural).
- − Anatomical issues with the subject's right hand holding the microphone.
Verdict: FLUX.2 [flex] does a better job of fulfilling the specific quantities and themes of the prompt by including several dogs in jerseys and a full newsroom set, though it loses some of the subject's specific facial features. Reve Image 1.0 captures a near-perfect caricature likeness of the woman and features a creative costume mashup, but it has more technical artifacts like the floating hockey stick and a malformed hand. FLUX.2 [flex] is the preferred winner for its more complete and coherent scene composition.
Studio Ghibli Anime Style
Editing“Transform this photo into a Studio Ghibli–inspired illustration. Use soft pastel colors, hand-painted textures, gentle lighting, dreamy backgrounds, and a warm, nostalgic mood”
AI Judge Analysis
FLUX.2 [flex]
- + Excellent preservation of the source image's composition and poses.
- + Successful application of a paper-like texture and soft pastel watercolor aesthetic.
- + High resolution and clean linework consistently applied across the frame.
- − The characters' emotional expressions are altered; the girlfriend now looks happy rather than jealous.
- − The man's expression is more surprised/confused than mesmerized.
Reve Image 1.0
- + Stronger adherence to the Studio Ghibli character design style, specifically in the eye shapes and shading.
- + Kept the girlfriend's angry/shocked expression, which is central to the original meme.
- − The foreground character is extremely blurry, losing the distinct focal balance of the original.
- − The background architectural details are cluttered and less coherent than Image A.
Verdict: FLUX.2 [flex] produced a much cleaner and more professional-looking illustration that perfectly preserved the layout and perspective of the original photo. While Reve Image 1.0 captured the specific Ghibli character aesthetic and the emotional narrative better, its technical execution suffered from excessive blurriness and less refined detail.
Golden Hour Stroll
Image Editing“Add dynamic motion to this photo: make hair blow in the wind, add leaves flying, energetic and lively feel.”
AI Judge Analysis
FLUX.2 [flex]
- + Successfully added wind-blown hair while keeping the face largely intact
- + Added leaves that feel grounded in the scene
- + Preserved the identity of the person and dog very accurately
- − The dog's tail pose changed unnecessarily to a more upright position
- − The blowing leaves are small and relatively sparse, lacking a strong dynamic punch
Reve Image 1.0
- + Excellent depiction of wind-blown hair with a strong sense of direction
- + Dynamic use of leaves with varying sizes and motion blur to create depth
- + High energetic and lively feel as requested in the prompt
- − Minor artifacting where some leaves overlap with the background elements
- − The hair edit slightly alters the silhouette of the head compared to the original
Verdict: Reve Image 1.0 is the winner because it interpreted 'dynamic' and 'lively' much more effectively than FLUX.2 [flex]. While both models handled the hair and leaves well, Reve Image 1.0 used motion blur and scale on the leaves to create a true sense of movement, whereas FLUX.2 [flex] felt more static.
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
FLUX.2 [flex]
- + Excellent typography with perfect spelling of 'APOLLO 11'.
- + Clean, modern vector aesthetic that aligns perfectly with the 'NASA-inspired' request.
- + Accurate and high-quality iconography for the Saturn V and Lunar Module.
- − Missed the final 'Landing' step, stopping at 'Descent'.
- − The layout flow is a bit non-linear, jumping from right to left.
Reve Image 1.0
- + Correctly included all requested steps through to the landing.
- + Includes a nice creative addition of the crew names.
- + Good use of numbering to guide the eye through the process.
- − Significant text rendering issues with multiple misspellings like 'Anmlot', 'Lunnr', and 'Dessent'.
- − Icons are slightly less polished and ‘crisp’ compared to Model A.
Verdict: FLUX.2 [flex] produced a much more professional and aesthetically pleasing infographic with perfect text rendering, though it missed the final step of the mission. Reve Image 1.0 followed the structural instructions more closely by including the landing, but failed significantly on text legibility and spelling. FLUX.2 [flex] is the preferred choice for its high-quality vector execution.
Explore each model
Reve AI's text-to-image generation model with strong aesthetic quality, accurate text rendering, and detailed instruction following capabilities