Black Forest Labs' 12-billion parameter multimodal flow transformer for in-context image generation and editing with character consistency, typography handling, and commercial-ready quality
Settled by community votes across 4 shared challenges, with an AI judge weighing in on each.
FLUX.1 Kontext [pro]
20.3
arena score
#41 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.1 Kontext [pro]
0%
win rate
Ties
0%
Qwen Image Edit 2511
0%
win rate
Shared challenges
4
Challenge by challenge
The strongest take from each model on every shared challenge, with the AI judge's read.
Bald man challenge
Image EditingEdit instruction
“Give the person a full, thick head of natural hair with realistic texture, density, and a natural hairline. Preserve facial features and lighting.”
FLUX.1 Kontext [pro]
Qwen Image Edit 2511
AI Judge Analysis
FLUX.1 Kontext [pro]
- + Natural integration of the hairline into the forehead skin
- + Excellent preservation of the jacket and background details
- − Subtly changed the shape of the glasses frame and nose
- − The hair texture is slightly stiff and lacks fine stray hairs
Qwen Image Edit 2511
- + Excellent preservation of the original facial anatomy and glasses
- + Realistic curly hair texture with natural-looking flyaways
- + Great preservation of the background and lighting
- − The hairline on the forehead looks a bit heavy or drawn-on compared to the scalp
Verdict: Qwen Image Edit 2511 is the preferred model because it managed to add a complex head of hair while perfectly preserving the person's identity, glasses, and specific facial features from the source. FLUX.1 Kontext [pro] also performed well but introduced subtle changes to the shape of the nose and eyewear that altered the subject's likeness.
Over-the-top cartoon caricature
EditingEdit 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.1 Kontext [pro]
Qwen Image Edit 2511
AI Judge Analysis
FLUX.1 Kontext [pro]
- + Excellent preservation of the source subject's hairstyle and clothing.
- + Successfully translates the image into a clean, stylized cartoon/caricature.
- − Completely ignored all functional instructions regarding the job (tv anchor), dogs, and hockey.
- − No exaggeration or humor added beyond the basic art style change.
Qwen Image Edit 2511
- + Successfully incorporated the tv anchor role, a dog, and hockey elements (hockey boards in the background and logo).
- + Applied a true caricature style with an exaggerated head and facial features.
- + Captures the humorous and busy energy requested in the prompt.
- − The text in the speech bubble is nonsensical.
- − An odd candle detail is included on the desk without clear reason.
Verdict: While FLUX.1 Kontext [pro] did a better job of preserving the original clothing and subtle facial likeness of the source image, it completely failed to follow the creative instructions of the prompt. Qwen Image Edit 2511 followed every instruction, successfully blending the subject's career as an anchor with her interests in dogs and hockey, resulting in a much more effective caricature.
Studio Ghibli Anime Style
EditingEdit 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.1 Kontext [pro]
Qwen Image Edit 2511
AI Judge Analysis
FLUX.1 Kontext [pro]
- + Excellent adherence to the Ghibli art style with watercolor-like textures and simplified facial features.
- + Accurately replicates the poses, clothing details, and compositions of the original meme in an illustrative format.
- + Captures the soft, warm lighting and nostalgic mood requested in the prompt.
- − The faces, while stylized, lose some of the specific intensity of the original expressions, especially the girlfriend's anger.
Qwen Image Edit 2511
- + Maintains high facial similarity and expression accuracy to the original source photo.
- + Successfully applies a soft, warm color palette and glow effects to the scene.
- − The illustration style feels more like a generic digital painting or 'Disney-fied' filter rather than Studio Ghibli's hand-drawn aesthetic.
- − Includes some small white artifacts/dots scattered across the image that weren't requested.
Verdict: FLUX.1 Kontext [pro] is the clear winner as it successfully transforms the image into a genuine Studio Ghibli art style with hand-painted watercolor textures and distinct line work. Qwen Image Edit 2511 applies a nice warm filter and softens the image, but the resulting style feels like a standard digital illustration rather than the specific Ghibli aesthetic requested.
Golden Hour Stroll
Image EditingEdit instruction
“Add dynamic motion to this photo: make hair blow in the wind, add leaves flying, energetic and lively feel.”
FLUX.1 Kontext [pro]
Qwen Image Edit 2511
AI Judge Analysis
FLUX.1 Kontext [pro]
- + Excellent source preservation, maintaining the woman's facial features almost perfectly.
- + Dynamic hair motion looks natural and realistic.
- + Successfully added falling leaves while keeping the background consistent.
- − The leaf placement is somewhat sparse.
- − The dog's tail lost some detail compared to the original.
Qwen Image Edit 2511
- + Highly energetic feel with a significant amount of flying leaves.
- + Clearly followed the instruction to make hair blow in the wind with a dramatic effect.
- + Good preservation of the overall scene layout.
- − The woman's face underwent slight changes that make her look different from the source image.
- − The hair strands look somewhat stylized/liquid rather than natural hair.
- − The dog's head shape and eyes were slightly altered.
Verdict: Both models successfully interpreted the dynamic motion request, but FLUX.1 Kontext [pro] is the better edit because it preserved the subject's identity much more accurately. While Qwen Image Edit 2511 provided more leaves and a higher sense of 'energy,' it inadvertently modified the facial features of the woman and the dog, whereas FLUX.1 Kontext [pro] kept the source image's details intact while adding the requested effects.
Next steps
Explore each model
Alibaba's Qwen image editing model for instruction-based image modifications and transformations