FLUX.2 [dev] Flash vs FLUX.2 [pro]
Head-to-head across 6 challenges
FLUX.2 [dev] Flash
60.0%
win rate
Ties
0.0%
FLUX.2 [pro]
40.0%
win rate
Challenge Results
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 [dev] Flash
- + Excellent material physics, showing realistic thick glass edges and reflections of the sphere on the bottom pane.
- + Highly detailed textures on the book cover and the wooden table surface.
- + Perfect adherence to all prompt elements including placement and lighting.
- − The glass has slight imperfections/dust specks which, while realistic, might look 'dirty' to some users.
FLUX.2 [pro]
- + Clean, minimalist aesthetic with very sharp, clear glass edges.
- + Good composition with a nice shallow depth of field on the background plant.
- + Accurately follows all positioning instructions in the prompt.
- − The blue sphere looks very matte and lacks the realistic caustic reflections or floor reflections seen in Model A.
- − The glass cube lacks the physical mass and refractive properties of real glass, looking more like thin acrylic.
Verdict: Both models followed the prompt perfectly. FLUX.2 [dev] Flash produced a significantly more realistic image with convincing glass thickness, complex reflections, and tactile textures on the wood and book. FLUX.2 [pro] is a very clean image but feels more like a 3D render due to the matte sphere and the lack of bottom-pane reflections.
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
FLUX.2 [dev] Flash
- + Excellent motion blur on the passing cars in the background
- + Very realistic skin texture and age detailing on the subject
- + Strong adherence to the 'imperfect framing' prompt with a centered, street-photography aesthetic
- − Physical logic errors with the bicycle spokes and frame intersecting the man's legs
- − The tools on the ground appear slightly muddy and poorly defined
FLUX.2 [pro]
- + Superior composition and use of shallow depth of field
- + More natural crouching pose for a bike repair context
- + Clean, high-quality rendering with fewer geometric artifacts in the bicycle
- − Lacks the requested 'motion blur from passing cars' as the background vehicles are static/sharp
- − Skin texture on the hands is slightly smoother/less detailed than the face
Verdict: FLUX.2 [dev] Flash captured the motion blur and 'imperfect' street photography feel much better, though it struggled with the physical coherence of the bicycle spokes. FLUX.2 [pro] produced a cleaner, more aesthetically pleasing image with better lighting, but completely missed the motion blur requirement. FLUX.2 [dev] Flash is the winner for better following the technical constraints of the prompt.
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 [dev] Flash
- + Excellent rendering of ornate engravings and fine metallic textures.
- + Highly detailed leather straps with realistic stitching and buckle hardware.
- + Perfect adherence to the 'hair braided with small beads' requirement with clear, distinct braids.
- − Symmetry in the background torches feels a bit artificial.
- − The character's expression is somewhat static compared to the battle-worn theme.
FLUX.2 [pro]
- + Superior skin texture and realistic depth in the scarring and dirt application.
- + Wonderful warm torchlight reflection that creates a more cohesive, atmospheric mood.
- + Dynamic and natural-looking bokeh sparks that blend well with the scene.
- − The leather straps appearing on the shoulder look slightly distorted or melted into the fabric.
- − The hair braids are a bit messy and less defined than in the other model.
Verdict: Both models followed the prompt exceptionally well, but FLUX.2 [pro] captures a more cinematic and realistic atmosphere with superior skin detailing and lighting integration. While FLUX.2 [dev] Flash has cleaner technical execution on the armor engravings and leather accessories, FLUX.2 [pro] feels more like a lifelike portrait due to its authentic handling of shadows and texture.
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
FLUX.2 [dev] Flash
- + Includes all four requested animals (dog, cat, bunny, fox).
- + Excellent rendering of light rays and dew sparkles in the meadow.
- + Very clear, expressive eyes on all subjects.
- − Includes an extra bunny, totaling five animals instead of four.
FLUX.2 [pro]
- + Highly dynamic and playful posing with animals 'tumbling' as requested.
- + Superb fur detail and lighting coherence on the puppy.
- + Good integration of butterflies into the interaction.
- − Completely missed the 'baby bunny' from the prompt.
- − Anatomical weirdness on the puppy's paws (extra toes/pads).
Verdict: FLUX.2 [dev] Flash adhered better to the prompt by including all four requested animal types, whereas FLUX.2 [pro] missed the bunny entirely. Although FLUX.2 [pro] had a more dynamic 'tumbling' composition, FLUX.2 [dev] Flash's overall adherence and clean rendering of the environment make it the better match for the specific request.
Heroic Super Hero Portrait
Text-to-Image“Hyper-photorealistic full-body portrait of a female superhero standing triumphantly on a New York skyscraper rooftop at golden sunset, wearing a classic modest superhero costume with flowing cape, chest emblem, gloves, and boots in red and blue colors, practical design, short hair, strong determined heroic expression looking into the distance, powerful confident stance with hands on hips and cape billowing dramatically in the wind, detailed urban cityscape background, warm natural sunlight with sharp shadows and fabric highlights, ultra-sharp textures on suit, hair, and concrete, 8K masterpiece, empowering family-friendly style.”
AI Judge Analysis
FLUX.2 [dev] Flash
- + Natural lighting and skin tones
- + Clear, high-quality urban background
- + Balanced composition that emphasizes the character's height
- − The fabric texture on the suit is slightly less detailed than Model B
- − Hands on hips look slightly stiff
FLUX.2 [pro]
- + Excellent intricate detailing on the armor and costume
- + Dynamic lighting with strong highlights on the suit
- + Solid posture that conveys the 'powerful confident stance' requested
- − The transition between the boot and leg on the left feels slightly awkward
- − The lighting on the face is a bit flatter compared to the environment
Verdict: Both models followed the prompt exceptionally well, capturing the golden hour NYC rooftop setting and specific costume requirements. FLUX.2 [pro] is the winner due to the superior complexity and texture of the suit design and the more dramatic billowing of the cape, which better fulfills the 'hyper-photorealistic' and 'heroic' aspects of the prompt.
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 [dev] Flash
- + Successfully includes two Earth and two Moon graphics to represent different stages.
- + Maintains high-detail renderings of the Saturn V and Lunar Module.
- + Includes the requested astronaut names correctly.
- − The layout is cluttered with overlapping elements and repetitive labels.
- − Includes hallucinated text like 'Sataurr Iccon' and multiple 'Landing' labels.
- − Visual style is more illustrative than the requested 'flat-vector' style.
FLUX.2 [pro]
- + Excellent clean, modern vector layout with a logical top-to-bottom flow.
- + Perfectly captures the 'flat-vector' style with crisp lines and consistent iconography.
- + Source-accurate NASA-inspired color palette and minimalist design.
- − The placeholder descriptive text is nonsensical gibberish.
- − The Saturn V rocket icon is a bit generic compared to the specific model requested.
Verdict: FLUX.2 [pro] (Model B) is the clear winner as it perfectly understands the 'infographic' and 'flat-vector' style requirements, creating a professional and balanced layout that reads like a real poster. While it uses gibberish for small body text, its overall aesthetic and composition are superior to FLUX.2 [dev] Flash (Model A), which is cluttered and struggles with consistent text placement.
FLUX.2 [dev] Flash
Fast distilled version of Black Forest Labs' FLUX.2 [dev] optimized for speed and cost efficiency.
FLUX.2 [pro]
Black Forest Labs' state-of-the-art image generation model with maximum quality and speed, supporting text-to-image and multi-reference image editing with up to 4MP output