Head to head
Esc

Models · slot A

to navigate to pick

FLUX.2 [klein] 4B Black Forest Labs Qwen Image Edit 2511 Alibaba

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

FLUX.2 [klein] 4B

22.3 arena score

#32 of 62 in Text-to-Image

Skill signature · Text-to-Image

Qwen Image Edit 2511

20.2 arena score

#16 of 32 in Image Editing

Vote tally

Where the votes landed

FLUX.2 [klein] 4B

0%

win rate

Ties

0%

Qwen Image Edit 2511

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] 4B
Qwen Image Edit 2511

AI Judge Analysis

FLUX.2 [klein] 4B

  • + Excellent preservation of the specific white Rolls Royce Phantom Drophead Coupé from the source image
  • + Accurate depiction of the California coastline with motion blur on the wheels
  • + Natural integration of the character's clothing texture into the seating position
  • The man's face is less recognizable compared to the source image
  • The composition is a wide exterior shot, making the driver a small detail

Qwen Image Edit 2511

  • + Superb face and hair preservation of the man from the source image
  • + Perfect translation of the specific plaid coat and black scarf into the new scene
  • + Dynamic and high-quality interior composition that captures the requested coastline perfectly
  • The car model was changed from a Rolls Royce to a generic luxury convertible
  • Minor steering wheel/arm anatomical clipping

Verdict: FLUX.2 [klein] 4B followed the prompt by keeping the exact car from the source but sacrificed the man's facial likeness. Qwen Image Edit 2511 completely changed the car model, but perfectly preserved the man's face, hair, and clothing, while creating a much more engaging 'lifestyle' photograph. Qwen is the winner for its superior ability to preserve the subject's identity in the edit.

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] 4B
Qwen Image Edit 2511

AI Judge Analysis

FLUX.2 [klein] 4B

  • + Successfully replicates the complex pose and environment of Image 1.
  • + Accurately transfers clothing details like the scarf and black sweatshirt from Image 2.
  • + Maintains character identity including the sunglasses and facial features.
  • The hair length is significantly longer than the character in Image 2.
  • The hand on the far right has anatomical issues and red nail polish that wasn't removed.

Qwen Image Edit 2511

  • + Keeps both characters in the frame, but fails the specific instruction of 'recreating' the person from Image 2 in the pose.
  • Fails the core instruction to put the character from Image 2 into the pose of Image 1.
  • Creates a nonsensical composition by stacking the images rather than performing an edit.
  • The feet of the foreground character are poorly integrated into the red stool.

Verdict: FLUX.2 [klein] 4B successfully understood the complex spatial task, placing the character from Image 2 into the dynamic pose of Image 1 while maintaining the background and lighting. Qwen Image Edit 2511 failed the prompt entirely, merely overlaying a cutout of the first image onto the second without merging the character identity or pose as requested.

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] 4B
Qwen Image Edit 2511

AI Judge Analysis

FLUX.2 [klein] 4B

  • + Perfectly preserved the subject's face, skin condition, and hair from Image 1
  • + Accurately replicated the plaid scarf and navy pea coat
  • + Effectively blended the subject's vitiligo-patterned skin visible through the open coat
  • The coat is slightly darker than the reference
  • A few extra gold chains were added that were not in Image 2

Qwen Image Edit 2511

  • + Successfully captured the blue hue of the coat from Image 2
  • + Included the sunglasses from Image 2
  • Completely failed to preserve the person from Image 1, replacing him with the man from Image 2
  • The scarf pattern is distorted and less accurate than Model A

Verdict: FLUX.2 [klein] 4B followed the instructions almost perfectly, successfully dressing the specific individual from Image 1 in the attire from Image 2 while maintaining his identity and unique skin features. Qwen Image Edit 2511 failed the core task of identity preservation, essentially swapping the person entirely for the man in the second image.

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] 4B
Before After
Qwen Image Edit 2511

AI Judge Analysis

FLUX.2 [klein] 4B

  • + Excellent preservation of the original facial features and expression.
  • + Natural, wavy hair texture that integrates well with the original beard.
  • + Perfectly preserves the background, clothing, and lighting conditions.

Qwen Image Edit 2511

  • + Generates a thick, curly head of hair that follows the prompt.
  • + Maintains the overall composition and lighting of the scene.
  • Significantly alters the forehead shape and eye area, making the person look different.
  • Visible artifacting where the hair meets the forehead.
  • The hair texture appears slightly more synthetic/plastic compared to Model A.

Verdict: FLUX.2 [klein] 4B is the clear winner because it successfully added the hair while keeping the subject's identity perfectly intact. Qwen Image Edit 2511 struggled with source preservation, noticeably changing the subject's facial structure and forehead in the process of adding the hair.

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] 4B
Qwen Image Edit 2511

AI Judge Analysis

FLUX.2 [klein] 4B

  • + Successfully incorporates all prompt elements: TV anchor desk, two dogs, and hockey players on the background screens.
  • + The facial features in the caricature clearly resemble the subject from the source image.
  • + High-quality line work and professional comic-style coloring.

Qwen Image Edit 2511

  • + Strong 'caricature' aesthetic with exaggerated expressions and features.
  • + Includes the dog and TV newsroom setting as requested.
  • + Good preservation of the subject's hair color and eye shape into the stylized version.
  • Completely missed the 'hockey' requirement in the prompt.
  • Contains nonsensical text in the speech bubble and strange artifacts like a candle on a news desk.

Verdict: FLUX.2 [klein] 4B followed the prompt much more accurately by including hockey-related imagery on the studio screens, whereas Qwen Image Edit 2511 ignored the hockey portion of the request entirely. While both models created effective caricatures with facial resemblance, FLUX.2 [klein] 4B's composition is more logical and detail-rich.

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] 4B
Qwen Image Edit 2511

AI Judge Analysis

FLUX.2 [klein] 4B

  • + Excellent adherence to the Studio Ghibli aesthetic with soft watercolor textures and line art.
  • + Successfully captures the requested 'dreamy' and 'warm, nostalgic' mood through pastel tones.
  • + High creative merit in reimagining the background into a lush, Ghibli-esque garden setting.
  • Loses the specific facial expressions of the original meme, particularly the girlfriend's look of indignation.
  • The character design is a more generic anime style rather than preserving the specific faces of the people in the photo.

Qwen Image Edit 2511

  • + Strongly preserves the original composition, including the exact facial expressions and background elements.
  • + Maintains the likeness of the people in the original photograph very effectively.
  • + Applies a cel-shaded illustration style that bridges the gap between the photo and an anime.
  • The lighting and colors feel more like a digital filter than the requested 'hand-painted' Ghibli texture.
  • The background remains a blurry urban street, missing the 'dreamy' quality requested in the prompt.

Verdict: FLUX.2 [klein] 4B follows the stylistic prompt much more accurately, delivering a beautiful watercolor-style illustration that feels like genuine Studio Ghibli concept art, though it loses the 'meme' facial tension. Qwen Image Edit 2511 is much better at preserving the source image's identity and expressions, but its artistic execution is less faithful to the specific Ghibli aesthetic requested. FLUX.2 is the preferred winner for better capturing the specific atmosphere and artistic style of the edit instruction.

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] 4B
Before After
Qwen Image Edit 2511

AI Judge Analysis

FLUX.2 [klein] 4B

  • + Excellent preservation of the source image's identity and facial features
  • + Very high density of flying leaves that feel integrated into the scene
  • + Realistic wind effect on hair that flows naturally in one direction
  • The leash handle has been slightly simplified compared to the source

Qwen Image Edit 2511

  • + Creates a very dramatic, symmetrical 'hair blow' effect
  • + Good variety in the colors of the flying leaves
  • + Maintains high structural consistency with the source image
  • The hair effect looks slightly artificial/stylized rather than wind-blown
  • The dog's tail has a slightly different shape/texture compared to the source

Verdict: Both models successfully interpreted the dynamic motion request. FLUX.2 [klein] 4B is preferred as its hair physics feel more grounded in reality and the high volume of leaves creates a stronger sense of a gusty day, whereas Qwen Image Edit 2511 opted for a more stylized, almost gravity-defying hair effect.

Next steps

Explore each model