Black Forest Labs' distilled 9 billion parameter image generation model with sub-second inference and multi-reference support
Settled by community votes across 4 shared challenges, with an AI judge weighing in on each.
FLUX.2 [klein] 9B
21.3
arena score
#7 of 23 in Image Editing
Skill signature
Not enough comparable category data
The chart appears once both models have ratings across at least three shared arena categories.
GPT Image 1.5
26.5
arena score
#7 of 44 in Text-to-Image
Top 3 in Image Editing
Vote tally
Where the votes landed
FLUX.2 [klein] 9B
50.0%
win rate
Ties
0.0%
GPT Image 1.5
50.0%
win rate
50.0%
0.0% ties
50.0%
Shared challenges
4
Challenge by challenge
The strongest take from each model on every shared challenge, with the AI judge's read.
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.”
FLUX.2 [klein] 9B
GPT Image 1.5
0% wins
0% ties
100% wins
AI Judge Analysis
FLUX.2 [klein] 9B
- + Excellent typography rendering and clean starburst icon.
- + High photorealistic quality with clear, appetizing textures on the bun and meat.
- + Modern professional composition suitable for a high-end food advertisement.
- − The 'exploded' effect is very conservative, with layers barely separated.
- − The background is more static and less 'fiery' compared to the other model.
GPT Image 1.5
- + Captures the 'exploded' request perfectly with wide separation of all ingredients.
- + Intense, high-energy background with dynamic embers and smoke.
- + Good text integration with a fiery glowing effect on the main title.
- − The 'LIMITED TIME ONLY' text is placed at the very bottom and slightly cut off by the frame.
- − Image has a slightly over-processed or 'deep fried' HDR look that reduces realism.
Verdict: FLUX.2 [klein] 9B produces a cleaner, more professional-looking advertisement with superior font rendering, but it plays it very safe with the 'exploded' burger concept. GPT Image 1.5 follows the core instruction of an exploded burger more literally and creates a more energetic atmosphere, though the text placement and image texture are less refined. FLUX.2 is preferred for overall design quality and realism.
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.”
FLUX.2 [klein] 9B
GPT Image 1.5
AI Judge Analysis
FLUX.2 [klein] 9B
- + Excellent chalk texture on the board surface
- + Highly accurate text rendering of the complex prompt
- + Strong aesthetic with realistic cursive and layout
- − Small spelling error in 'fress' instead of 'fresh'
- − Composition feels slightly crowded with the smudges
GPT Image 1.5
- + Clean layout with consistent handwriting
- + Accurate spelling throughout the entire text
- + Realistic wooden frame integration
- − Texture appears more like a digital overlay than physical chalk
- − The cursive title is less elegant compared to the competing model
Verdict: Both models followed the complex prompt remarkably well, including the specific date and menu prices. FLUX.2 [klein] 9B is the preferred choice due to its superior chalk texture and more authentic handwriting variations, despite a minor typo. GPT Image 1.5 is visually cleaner but lacks the tactile, dusty quality of a real chalkboard.
Pose & Character Mashup
EditingEdit instruction
“Use Image 1 as the exact pose reference and Image 2 as the character reference. Recreate the person/character from Image 2 in the exact dynamic pose and body position from Image 1. Keep the exact face, hair, clothing style/details, and expression from Image 2. Match the lighting and environment of Image 1. The final image must show the character from Image 2 performing the precise action/pose from Image 1 with perfect anatomy and natural integration.”
Source
FLUX.2 [klein] 9B
GPT Image 1.5
100% wins
0% ties
0% wins
AI Judge Analysis
FLUX.2 [klein] 9B
- + Near-perfect adherence to the complex anatomical pose including head tilt and arm angle.
- + Highly accurate recreation of character accessories like the scarf pattern and sunglasses.
- + Excellent lighting consistency with the yellow studio environment.
- − One hand is clenched into a fist, differing from the reference pose's open fingers.
- − Minor skin tone inconsistency between the face and feet.
GPT Image 1.5
- + Faithful reproduction of the character's facial features and specific expression.
- + Good preservation of the red box prop from the source environment.
- − Fails the core pose requirement by keeping the torso and head upright.
- − Proportions are slightly distorted where the legs cross.
Verdict: FLUX.2 [klein] 9B is the clear winner as it successfully replicates the difficult, dynamic body position and head tilt from the pose reference. GPT Image 1.5 fails to capture the lean and specific orientation of the original image, resulting in a much more static composition.
Outfit Transfer Challenge
EditingEdit instruction
“Use Image 1 as the base person. Dress them in the exact elaborate outfit from Image 2 (including all layers, accessories, jewelry, and shoes). Carefully adapt the clothing to the body shape and pose in Image 1 while maintaining realistic fabric behavior, correct proportions, and perfect lighting/shadow matching. Keep the person’s exact face, hair, and background completely unchanged.”
Source
FLUX.2 [klein] 9B
GPT Image 1.5
AI Judge Analysis
FLUX.2 [klein] 9B
- + Successfully preserved the subject's face, hair, and vitiligo patterns
- + Accurately replicated the coat, scarf, and jeans from the reference image
- + Included the gold jewelry and watch as requested
- − Lighting on the coat is slightly too bright compared to the beach environment
GPT Image 1.5
- + Perfectly replicated the texture and drape of the plaid scarf
- + Maintained the background and wooden structure accurately
- − Failed to include the subject's face, which was a core instruction
- − Missing the gold jewelry shown in the base outfit
Verdict: FLUX.2 [klein] 9B followed all instructions, including the critical requirement to keep the person's face unchanged while applying the complex outfit from the second image. GPT Image 1.5 failed by cropping out the subject's head entirely and omitting jewelry. FLUX.2 is the clear winner for maintaining character identity and outfit completeness.
Next steps
Explore each model
OpenAI's state-of-the-art image generation model with better instruction following and adherence to prompts