Black Forest Labs' open-weights image generation model with frontier performance, available for non-commercial local deployment
Settled by community votes across 4 shared challenges, with an AI judge weighing in on each.
FLUX.2 [dev]
24.5
arena score
#17 of 44 in Text-to-Image
Skill signature
Not enough comparable category data
The chart appears once both models have ratings across at least three shared arena categories.
Grok Imagine Image
24.1
arena score
#19 of 44 in Text-to-Image
Vote tally
Where the votes landed
FLUX.2 [dev]
71.4%
win rate
Ties
14.3%
Grok Imagine Image
14.3%
win rate
71.4%
14.3% ties
14.3%
Shared challenges
4
Challenge by challenge
The strongest take from each model on every shared challenge, with the AI judge's read.
Geometric Composition
Text-to-Image“A glass cube on a wooden table. Inside the cube is a small blue sphere. On top of the cube sits a red book. A green plant is behind the cube, partially visible through the glass. Soft window light from the left.”
FLUX.2 [dev]
Grok Imagine Image
0% wins
0% ties
100% wins
AI Judge Analysis
FLUX.2 [dev]
- + Excellent representation of thick glass with realistic internal reflections and corner details.
- + Physical placement of the sphere at the bottom of the cube feels more grounded and realistic.
- + Soft window light is beautifully rendered, creating a natural atmosphere.
- − The plant in the background is very blurry, making it less distinct than requested.
- − The sphere has a slight transparency that might not be expected for a simple 'blue sphere'.
Grok Imagine Image
- + The plant is clearly 'behind the cube' and visible through the glass as requested.
- + The cube edges are sharp and clean, creating a very modern aesthetic.
- + The red book has a nice texture on the cover and realistic page edges.
- − The blue sphere is levitating in the center of the cube, which feels physically unnatural despite the prompt not specifying it should be floating.
- − The reflections on the table surface are slightly disconnected from the objects.
Verdict: FLUX.2 [dev] produces a more photorealistic scene with superior lighting and material physics, particularly in how it handles the weight and reflections of the glass. While Grok Imagine Image followed the spatial instruction for the plant more clearly, the floating blue sphere and slightly thinner glass rendering make it feel less grounded than FLUX.2 [dev].
Candid Street Photography
Text-to-Image“A candid street photo of an elderly Japanese man repairing a red bicycle in light rain, reflections on wet pavement, shallow depth of field, 50mm lens, natural skin texture, imperfect framing, motion blur from passing cars, cinematic but realistic, no stylization.”
FLUX.2 [dev]
Grok Imagine Image
100% wins
0% ties
0% wins
AI Judge Analysis
FLUX.2 [dev]
- + Excellent depiction of motion blur in the background cars.
- + Highly detailed and realistic facial features and skin texture.
- + Strong composition that highlights the subject while maintaining the street atmosphere.
- − The red bicycle's frame and handlebars are structurally confusing and messy.
- − The rain effect is very subtle, making the pavement look more like it's already wet rather than currently raining.
Grok Imagine Image
- + Captured the 'imperfect framing' prompt well with a more candid, snapshot aesthetic.
- + The bicycle structure is much more coherent and realistic.
- + Good use of color and lighting for a cinematic feel.
- − The subject's face is obscured and lacks the 'natural skin texture' detail requested.
- − Motion blur on the background car is less convincing compared to Model A.
Verdict: FLUX.2 [dev] excels in photographic detail, particularly in the subject's face and the convincing motion blur of the passing traffic, though the physical structure of the bicycle is warped. Grok Imagine Image better captures the requested 'imperfect framing' and provides a more realistic bicycle, but it fails to deliver the high-quality skin textures and facial detail requested in the prompt. FLUX.2 [dev] is the winner for its superior rendering of the complex lighting and textures required for a realistic cinematic shot.
Fantasy Warrior
Text-to-Image“Close portrait of a battle-worn paladin in ornate engraved plate armor, hair braided with small beads, faint scars and dirt on the skin, warm torchlight reflecting off metal, shallow depth of field, bokeh sparks, lifelike eyes, highly detailed texture on leather straps and cloth underlayer.”
FLUX.2 [dev]
Grok Imagine Image
100% wins
0% ties
0% wins
AI Judge Analysis
FLUX.2 [dev]
- + Excellent execution of the leather straps and buckle textures
- + Accurate depiction of hair beads as requested in the prompt
- + More realistic skin texture with grime and faint scars
- − The scars look a bit like fresh cuts rather than faint, healed scars
- − Armor engraving is slightly less intricate than the competitor
Grok Imagine Image
- + Extremely intricate and beautiful armor engravings
- + Strong cinematic lighting with vibrant bokeh sparks
- + High contrast and sharp focus on the facial features
- − The skin looks a bit too smooth and airbrushed for a 'battle-worn' character
- − The 'hair beads' are less distinct, looking more like hair ties or wraps
- − Less visible texture on the leather straps compared to Model A
Verdict: Both models followed the prompt very well, but FLUX.2 [dev] stands out for its superior textural realism, particularly on the leather and skin, as well as the more literal interpretation of 'hair beads'. Grok Imagine Image produced a very aesthetically pleasing image with more ornate armor, but the face looks slightly too clean and stylized for the 'battle-worn' description.
Adorable Baby Animals in Sunny Meadow
Text-to-Image“Hyper-photorealistic scene of fluffy baby animals—a golden retriever puppy, tabby kitten, baby bunny, and red fox kit—with big expressive eyes and ultra-detailed soft fur, playfully chasing butterflies and tumbling together in a lush wildflower meadow, warm golden sunrise light with god rays and dew sparkles, joyful wholesome vibe, 8K masterpiece.”
FLUX.2 [dev]
Grok Imagine Image
75% wins
25% ties
0% wins
AI Judge Analysis
FLUX.2 [dev]
- + Excellent fur texture and realistic lighting interactions.
- + Natural composition with varied, lifelike poses for the animals.
- + Very high prompt adherence including the specifically requested animals and butterflies.
- − Included two rabbits instead of the requested one baby bunny.
- − The animals are mostly sitting rather than 'tumbling together'.
Grok Imagine Image
- + Bold, vibrant colors with distinct god rays.
- + Cute, expressive faces that lean into the 'wholesome vibe'.
- − Anatomical issues with the kitten, particularly the paws and flat facial structure.
- − The 'butterflies' are rendered more like small insects or moths.
- − Less photorealistic and more stylized/CGI-like compared to the prompt's request.
Verdict: FLUX.2 [dev] produces a significantly more realistic image with superior fur detail and natural lighting, though it added an extra rabbit. Grok Imagine Image has a more saturated, 'cuter' aesthetic but suffers from anatomical distortions and fails to render convincing butterflies. FLUX.2 [dev] is the clear winner for its adherence to the 'hyper-photorealistic' part of the prompt.
Next steps
Explore each model
An image generation model by xAI designed to generate highly aesthetic images from text descriptions.