Head to head
Esc

Models · slot A

to navigate to pick

FLUX.2 [klein] 9B Black Forest Labs Qwen Image Edit 2509 Alibaba

Settled by community votes across 7 shared challenges, with an AI judge weighing in on each.

FLUX.2 [klein] 9B

20.6 arena score

#13 of 32 in Image Editing

Skill signature · Image Editing

Qwen Image Edit 2509

7.8 arena score

#32 of 32 in Image Editing

Vote tally

Where the votes landed

FLUX.2 [klein] 9B

0%

win rate

Ties

0%

Qwen Image Edit 2509

0%

win rate

Shared challenges 7

Challenge by challenge

The strongest take from each model on every shared challenge, with the AI judge's read.

Man and Car in California

Editing
Edit instruction

“Make a photo of the man driving the car down the California coastline”

Source
FLUX.2 [klein] 9B
Qwen Image Edit 2509

AI Judge Analysis

FLUX.2 [klein] 9B

  • + Excellent preservation of the specific car model's chassis and details
  • + Realistic motion blur on the wheels and road
  • + Good integration of the man's hair texture into the driver seat
  • The driver's face is obscured by the car's pillar
  • The man's hair seems slightly detached from his head

Qwen Image Edit 2509

  • + Successfully shows the man's face and hair in the driver's seat
  • + Beautiful, cinematic lighting and scenery
  • + Maintains the car's proportions and brand identity well
  • The man's hair is blowing in the opposite direction of the car's motion
  • Loss of some fine details on the car front compared to original

Verdict: Both models performed remarkably well at combining two distinct source images into a new scene. FLUX.2 [klein] 9B preserved the car more accurately in terms of texture and lighting, but obscured the man's face. Qwen Image Edit 2509 is the winner for better composition, as it clearly shows the man driving while maintaining the California coastline aesthetic, despite a slight physics error with the hair direction.

Pose & Character Mashup

Editing
Edit 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
Qwen Image Edit 2509

AI Judge Analysis

FLUX.2 [klein] 9B

  • + Successfully replicates the complex pose from Image 1.
  • + Accurately applies the character's face, sunglasses, scarf, and clothing from Image 2.
  • + Matches the yellow studio lighting and red ottoman background perfectly.
  • The left foot (on the right side of the stool) has an extra toe or distorted anatomy.
  • The left hand is clenched in a fist rather than the open-handed pose from Image 1.

Qwen Image Edit 2509

  • + Maintains the pose and background of Image 1.
  • Completely failed to incorporate the character from Image 2.
  • Image appears to be a slightly filtered version of Image 1 with no character change.
  • The face, hair, and clothing are still those of the woman from the reference image.

Verdict: FLUX.2 [klein] 9B successfully followed the complex instructions, merging the character's identity from Image 2 with the dynamic pose from Image 1. Qwen Image Edit 2509 failed the task entirely, returning the original person from the pose reference with no character modification.

Outfit Transfer Challenge

Editing
Edit 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
Qwen Image Edit 2509

AI Judge Analysis

FLUX.2 [klein] 9B

  • + Excellent preservation of the subject's face, hair, and unique skin patterns.
  • + Highly realistic integration of the coat and scarf with believable lighting.
  • + Good background preservation.
  • Added significant jewelry (gold necklaces) that was not present in Image 2.
  • Did not render the full body/shoes as requested by the pose transition.

Qwen Image Edit 2509

  • + Successfully rendered the full outfit including boots as requested.
  • + Included the sunglasses from Image 2.
  • Failed to preserve the subject's face, merging it with the person from Image 2.
  • The lighting on the jeans and boots is very flat and looks poorly composited.
  • Overall image quality is lower with softer details than Model A.

Verdict: FLUX.2 [klein] 9B followed the instruction to keep the person's face and hair unchanged much better than Qwen Image Edit 2509, which created a hybrid person that lost the original subject's identity. While Qwen attempted the full-body edit including shoes, the execution was poor compared to the high-quality, realistic garment rendering and skin texture preservation in the FLUX.2 result.

Bald man challenge

Image Editing
Edit instruction

“Give the person a full, thick head of natural hair with realistic texture, density, and a natural hairline. Preserve facial features and lighting.”

Before After
FLUX.2 [klein] 9B
Before After
Qwen Image Edit 2509

AI Judge Analysis

FLUX.2 [klein] 9B

  • + Excellent texture and natural curly flow that matches the character's rugged aesthetic.
  • + Perfect preservation of the face, clothing, and background from the source image.
  • + The hairline integration with the sideburns across the glasses is very realistic.
  • One small stray hair curl near the top-center looks slightly detached from the main mass.

Qwen Image Edit 2509

  • + Very accurate preservation of the underlying person and background elements.
  • + Provides a thick, voluminous head of hair as requested.
  • The hair texture looks somewhat synthetic and 'plastic' with overly uniform waves.
  • The hairline above the forehead is a bit too straight and sharp to look fully natural.
  • The hair volume is slightly disproportionate to the head shape.

Verdict: FLUX.2 [klein] 9B is the winner because it provides a much more convincing and natural hair texture that fits the rugged look of the man in the photo. While Qwen Image Edit 2509 also preserves the original image well, its hair generation looks more like a wig due to the uniform, shiny waves and less realistic hairline.

Over-the-top cartoon caricature

Editing
Edit instruction

“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.”

Source
FLUX.2 [klein] 9B
Qwen Image Edit 2509

AI Judge Analysis

FLUX.2 [klein] 9B

  • + Successfully incorporates all prompt elements: tv anchor, dogs, and hockey stadium background.
  • + Better facial likeness preservation from the source image within the caricature style.
  • + High level of detail in the background crowd and studio equipment.
  • The caricature style is a bit busy and the hands are somewhat awkward.
  • Text on screens and signs is nonsensical gibberish.

Qwen Image Edit 2509

  • + Successfully preserves the clothing (denim shirt) from the source image.
  • + Captures the 'selfie' perspective of the original photo well.
  • + Includes a hockey rink and a dog as requested.
  • The facial caricature is less recognizable as the woman in the source photo.
  • The small dog is very simple and looks disconnected from the scene.
  • Speech bubble contains nonsensical text.

Verdict: FLUX.2 [klein] 9B is the superior model for this task as it successfully integrated all the requested themes—anchoring, hockey, and dogs—into a cohesive and visually rich scene that clearly resembles the subject. While Qwen Image Edit 2509 did a better job of preserving the specific clothing from the source image, its caricature of the face is less accurate and the overall composition feels more disjointed.

Studio Ghibli Anime Style

Editing
Edit instruction

“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”

Source
FLUX.2 [klein] 9B
Qwen Image Edit 2509

AI Judge Analysis

FLUX.2 [klein] 9B

  • + Excellent adherence to the 'Studio Ghibli' aesthetic with soft watercolor textures and line art.
  • + Successfully changes the environment to a dreamy pastoral setting common in Ghibli films.
  • + Preserves the composition and poses while effectively stylizing faces.
  • The change from a city to a field may be considered poor source preservation of the background.
  • Softens the man's expression to be more pleasant, losing the 'distracted' humor of the original meme.

Qwen Image Edit 2509

  • + Better preservation of the original city street background elements.
  • + Retains the man's specific facial expression and stubble more accurately to the source image.
  • + Correctly applies a soft anime/illustration overlay.
  • The texture feels more like a generic modern anime or digital halftone filter than Ghibli's hand-painted style.
  • The woman in the red dress has slightly distorted facial proportions compared to the original.

Verdict: FLUX.2 [klein] 9B is the winner because it successfully captures the specific 'Studio Ghibli' artistic style with beautiful watercolor textures and pastoral moods, whereas Qwen Image Edit 2509 looks more like a standard comic-style filter. While Qwen preserved the original city background better, FLUX.2 better fulfilled the 'dreamy' and 'nostalgic' atmosphere requested in the prompt.

Golden Hour Stroll

Image Editing
Edit instruction

“Add dynamic motion to this photo: make hair blow in the wind, add leaves flying, energetic and lively feel.”

Before After
FLUX.2 [klein] 9B
Before After
Qwen Image Edit 2509

AI Judge Analysis

FLUX.2 [klein] 9B

  • + Successfully added numerous falling leaves with varied depth and motion blur.
  • + The hair rendering looks natural and follows a consistent wind direction.
  • + Maintained high source preservation, keeping the person and background almost identical to the original.
  • Some leaves appear slightly repetitive in shape.

Qwen Image Edit 2509

  • + The hair has a very strong sense of motion and rhythmic flow.
  • + Increased the energetic feel by also adding motion to the dog's tail.
  • The added maple-style leaves look very flat and lack the photographic realism of the source image.
  • The hair texture has become slightly more stylized and less realistic compared to the original hair.

Verdict: Both models followed the instructions well, but FLUX.2 [klein] 9B is the superior edit because its additions—the leaves and the hair—blend seamlessly with the photographic style of the source. Qwen Image Edit 2509 added more dynamic elements like the dog's tail moving, but the leaves look like pasted-on clipart rather than part of the actual scene.

Next steps

Explore each model