Black Forest Labs' open-weights multimodal flow transformer for in-context image generation and editing, available for non-commercial use with character consistency and style transfer capabilities
Settled by community votes across 4 shared challenges, with an AI judge weighing in on each.
FLUX.1 Kontext [dev]
17.1
arena score
#54 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 [dev]
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 [dev]
Qwen Image Edit 2511
AI Judge Analysis
FLUX.1 Kontext [dev]
- + Successfully added thick hair with a modern style.
- + Maintains the general color scheme of the original clothing.
- − Significantly altered the facial structure, making the man look younger and like a different person.
- − The hairline looks slightly artificial and overly dense at the scalp.
- − Changed the frame and lens shape of the glasses.
Qwen Image Edit 2511
- + Excellent source preservation, keeping the original facial features almost perfectly intact.
- + Realistic, curly texture that blends naturally with the existing beard.
- + Maintains the original glasses and lighting conditions.
- − The hair volume is quite high, which might feel slightly exaggerated for some interpretations, though it fits the 'full, thick' prompt.
Verdict: Qwen Image Edit 2511 is the clear winner as it successfully adds the hair while preserving the identity of the person in the source image. FLUX.1 Kontext [dev] failed the 'preserve facial features' instruction, effectively generating a new face that only vaguely resembles the original.
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 [dev]
Qwen Image Edit 2511
AI Judge Analysis
FLUX.1 Kontext [dev]
- + Maintains the subject's original outfit (denim shirt) accurately.
- + Captures the user's facial likeness in a clean, comic-book style.
- + Includes a dog as requested.
- − Failed to include any hockey-related elements.
- − The 'TV anchor' profession is poorly represented by just a small television in the background.
- − Floating 'JOB' text is uninspired and literal.
Qwen Image Edit 2511
- + Excellent caricature style with exaggerated features and high-energy expression.
- + Perfectly captures the 'TV news anchor' setting with a studio background and microphone.
- + Successfully incorporates a dog into the composition.
- − Failed to include any hockey-related elements.
- − Completely changed the subject's outfit from the source image.
- − Contains odd artifacts like a candle on a laptop and nonsensical text.
Verdict: Both models failed to incorporate the 'hockey' element of the prompt. However, Qwen Image Edit 2511 is the superior caricature as it actually places the subject in a professional TV studio setting with an appropriately exaggerated facial expression, whereas FLUX.1 Kontext [dev] keeps the subject too static and fails to convey the requested profession effectively.
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 [dev]
Qwen Image Edit 2511
AI Judge Analysis
FLUX.1 Kontext [dev]
- + Excellent character translation into a clean anime aesthetic
- + Strong preservation of the original's layout and character poses
- + Consistent line art and vibrant colors
- − Style is more modern TV anime than the requested hand-painted Ghibli style
- − Atmospheric lighting and textures are missing
Qwen Image Edit 2511
- + Stronger adherence to the 'hand-painted textures' and 'dreamy background' prompt
- + Captures the soft, warm, nostalgic lighting associated with Studio Ghibli
- + Maintains superior facial likeness to the original source subjects
- − Includes some strange sparkling/dust artifacts across the image
- − Background architecture is slightly more washed out than necessary
Verdict: While FLUX.1 Kontext [dev] creates a very clean and attractive anime conversion, it feels more like modern digital animation than the specific Studio Ghibli style requested. Qwen Image Edit 2511 better captures the prompt's request for soft pastels, painterly textures, and a nostalgic mood, while also preserving the unique facial features of the people in the original meme much more effectively.
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 [dev]
Qwen Image Edit 2511
AI Judge Analysis
FLUX.1 Kontext [dev]
- + Excellent source preservation of the subject's face and clothing
- + Subtle and realistic hair motion modification
- + High consistency with the original background details
- − The flying leaves are very few and look like small green artifacts rather than natural foliage
- − The sense of 'energetic motion' is quite minimal
Qwen Image Edit 2511
- + Strong adherence to the request for many flying leaves
- + Significant and visually appealing hair motion that feels dynamic
- + Creates a much more lively and 'energetic' atmosphere as requested
- − Slight alteration of the facial features compared to the source image
- − Some leaves appear slightly blurred or inconsistent in lighting
Verdict: Qwen Image Edit 2511 followed the prompt's instructions much more effectively, providing dramatic hair movement and plenty of flying leaves to create an energetic feel. While FLUX.1 Kontext [dev] preserved the original subject's likeness better, it failed to meaningfully 'add' the requested dynamism, resulting in an edit that looks almost identical to the source.
Next steps
Explore each model
Alibaba's Qwen image editing model for instruction-based image modifications and transformations