Black Forest Labs' premium multimodal flow transformer with greatly improved prompt adherence and typography generation for in-context image generation and editing without compromise on speed
Settled by community votes across 4 shared challenges, with an AI judge weighing in on each.
FLUX.1 Kontext [max]
23.7
arena score
#23 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 [max]
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 [max]
Qwen Image Edit 2511
AI Judge Analysis
FLUX.1 Kontext [max]
- + Successfully added a thick head of hair with realistic volume.
- + Maintains the overall lighting and environmental atmosphere.
- − Significantly altered the facial features, making the man look like a different person.
- − Changed the color and style of the glasses frames.
- − Lowered the ear placement and changed its shape.
Qwen Image Edit 2511
- + Excellent preservation of the original facial features and glasses.
- + Very realistic hair texture and natural integration with the sideburns.
- + Perfectly maintains the background and clothing from the original image.
- − Slightly visible artifacting where the hair meets the forehead.
Verdict: Qwen Image Edit 2511 performed a near-perfect edit, successfully adding realistic curly hair while keeping the person's unique facial features and accessories identical to the source. FLUX.1 Kontext [max] failed at source preservation, generating a man who looks significantly different from the original subject and changing the style of his eyewear.
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 [max]
Qwen Image Edit 2511
AI Judge Analysis
FLUX.1 Kontext [max]
- + Maintains the subject's original denim jacket attire.
- + Captures a clear 'hockey' element with the stick in the background.
- + Stronger likeness preservation despite the caricature style.
- − The facial features feel more like a generic cartoon than a stylized exaggeration of the specific person.
- − The background news set is very minimal.
Qwen Image Edit 2511
- + Excellent 'exaggerated and humorous' caricature style with classic proportions.
- + Highly detailed TV news studio environment.
- + Good incorporation of the dog element with a cute, expressive character.
- − Changes the subject's clothing from casual denim to a formal suit and tie.
- − Completely misses the hockey requirement.
- − Includes nonsensical text in a speech bubble.
Verdict: FLUX.1 Kontext [max] followed the prompt more accurately by including the hockey element, which Qwen Image Edit 2511 missed entirely. While Qwen Image Edit 2511 has a more professionally polished 'caricature' art style, it failed to preserve the subject's original clothing and added gibberish text.
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 [max]
Qwen Image Edit 2511
AI Judge Analysis
FLUX.1 Kontext [max]
- + Excellent adherence to the Studio Ghibli art style with characteristic facial features and line work.
- + Perfectly captures the hand-painted watercolor texture and soft pastel palette requested.
- + Maintains the core composition and character roles of the original meme flawlessly.
- − The man's expression is slightly more neutral/sad than the original's 'distracted' look.
Qwen Image Edit 2511
- + Preserves the likeness and facial expressions of the original photo subjects very accurately.
- + Effective use of dreamy, warm lighting and golden color palette.
- − The art style is more generic 'manhwa/webtoon' than the requested Studio Ghibli style.
- − Includes distracting white speckled artifacts across the image that look like digital noise or dust.
Verdict: FLUX.1 Kontext [max] successfully transformed the image into a genuine Studio Ghibli-inspired illustration, capturing the specific hand-drawn aesthetic and textures requested. Qwen Image Edit 2511 preserved the original facial expressions and likenesses better, but failed to reach the specific artistic target, resulting in a more generic digital illustration with messy visual artifacts.
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 [max]
Qwen Image Edit 2511
AI Judge Analysis
FLUX.1 Kontext [max]
- + Natural integration of flying leaves into the background and foreground
- + Successfully added blowing hair while maintaining a realistic texture
- − Anatomical error in the left arm which appears detached or floating
- − Added a third hand/arm reaching toward the dog's head
Qwen Image Edit 2511
- + Excellent adherence to all prompts including flying leaves and wind-blown hair
- + Highly successful preservation of the original person and dog's identity and posture
- − The hair strands appear slightly stylized or 'painterly' compared to the source image
- − Some leaves appear slightly flat against the 2D plane
Verdict: Both models followed the instructions well, but FLUX.1 Kontext [max] introduced severe anatomical errors, including a floating limb and a nonsensical third hand. Qwen Image Edit 2511 accomplished the task while perfectly preserving the source subjects and maintaining a high level of physical coherence.
Next steps
Explore each model
Alibaba's Qwen image editing model for instruction-based image modifications and transformations