Midjourney v6 vs Adobe Firefly 2 vs Meta AI

Comparing the generation of photorealistic portraits

prompt: a young woman attending a music festival, backlighting, portrait

🥇 Midjourney (left) 🥈 Adobe Firefly (center) 🥈 Meta AI (right). Mijourney, a leading AI for image generation, demonstrates its artistic prowess and exceptional detail in this example. Although the generated image may appear darker initially, a closer look reveals its superior quality compared to the outputs of other neural networks. This showcases Mijourney's ability to produce highly detailed and visually striking images.

Links to images: Midjourney Adobe Firefly Meta AI

prompt: a physics professor teaching his class, academia, close-up

🥇 Midjourney (left) 🥈 Adobe (center) 🥉 Meta (right). Midjourney again secured the top spot for rendering quality. While Adobe Firefly produced a good result, some extraneous artifacts, like the 3D spheres on the blackboard, detracted from the overall realism. Meta's AI image generator, on the other hand, delivered the lowest quality output, which becomes evident upon closer inspection.

Links to images: Midjourney Adobe Firefly Meta AI

prompt: Martin Schoeller's close-up portrait of frontman of death-metal band

🥇 Midjourney (left) ❌ Adobe (center) 🥈 Meta (right). Midjourney once again produced the highest quality image with superior detail. Adobe Firefly failed to render a correct image.

Links to images: Midjourney Adobe Firefly Meta AI

prompt: Kourtney Roy's photographic portrait of old seafarer

🥇 Midjourney (left) 🥈 Adobe (center) 🥉 Meta (right). Adobe Firefly's generated images often exhibit a stock photo aesthetic, likely due to the model's training data which includes a significant portion of stock photography. However, this style may be suitable for certain cases. Meta's AI image generator, in this instance, produced an image with noticeable flaws in the rendering of fingers, a common challenge for modern neural networks.

Links to images: Midjourney Adobe Firefly Meta AI

prompt: a photo of cyberpunk post apocolyptic roadwarrior woman with shaved head ordering a drink at a dimly lit rough bar. she has abundant tatoos and peircings. leather vest, leather boots

🥇 Midjourney (left) 🥉 Adobe (center) 🥉 Meta (right). Midjourney takes the lead here. While Adobe's image boasts realism, it falls short in capturing the true essence of the prompt. Meta's image, on the other hand, lacks realism and suffers from those familiar rendering problems, especially with fingers.

Links to images: Midjourney Adobe Firefly Meta AI

prompt: close-up portrait of a weathered fisherman, wrinkles around his eyes, salt-spray on his beard, hyperrealistic textures, cinematic lighting

🥇 Midjourney (left) 🥈 Adobe (center) 🥇 Meta (right). While Midjourney once again delivered an image with exceptional detail, Meta's output, in this case, exhibited a higher degree of realism. Considering the importance of both detail and realism, this time we decided to award a joint first place to both Midjourney and Meta.

Links to images: Midjourney Adobe Firefly Meta AI

prompt: Serge Lutens's photographic portrait of young pirate-queen

🥇 Midjourney (left) 🥉 Adobe (center) 🥇 Meta (right). While the image generated by Adobe Firefly exhibits realism, it does not align with the specific requirements of the prompt. The depicted individual appears younger than the intended age for the 'pirate queen' character. Additionally, the overall aesthetic once again resembles typical stock photography.

Links to images: Midjourney Adobe Firefly Meta AI

prompt: street fashion photo, vibrant city murals as a backdrop, model in sleek modern wear with neon accents, reflective sunglasses capturing the city's pulse, candid pose

🥇 Midjourney (left) 🥈 Adobe (center) 🥈 Meta (right)

Links to images: Midjourney Adobe Firefly Meta AI

prompt: a groom wearing a light gray suit with a sage green tie and pocket square, glamour street medium format photography, portrait, natural lighting

🥇 Midjourney (left) 🥇 Adobe (center) 🥈 Meta (right). All AIs did a decent job, but Meta's image lacked the sharpness and detail of the others, earning it second place.

Links to images: Midjourney Adobe Firefly Meta AI

prompt: photorealistic photograph of a weathered farmer's face, sun-kissed skin framed by windswept hair

🥇 Midjourney (left) 🥉 Adobe (center) 🥇 Meta (right). Midjourney produced the image with the highest level of detail, while Meta's output excelled in realism. Adobe Firefly's result, however, deviated from the expectations set by the prompt. Additionally, it's worth noting a recurring trend in images generated by Adobe Firefly where subjects often exhibit smiles, regardless of the context.

Links to images: Midjourney Adobe Firefly Meta AI

prompt: last samurai. Cinematic portrait by Wong Kar-Wai

Links to images: Midjourney Adobe Firefly Meta AI

prompt: Ara Guler's portrait of 1970s Istanbul casino gambler

🥇 Midjourney (left) 🥉 Adobe (center) 🥉 Meta (right). Midjourney emerged as the frontrunner in this round. While the portrait of the man in Adobe Firefly's image demonstrates realism, the background fails to accurately depict 1970s Istanbul, and the overall aesthetic does not evoke the style of a 1970s portrait. Meta's output lacks realism, although it attempts to capture the essence of the prompt.

Links to images: Midjourney Adobe Firefly Meta AI

prompt: 1980s mystery film, low-angle shot of an evil-eyed French Butler sporting a black suit and grasping a candle in the hallway of a creepy Victorian mansion with musty decor. The warm candle glow evokes a spooky sense of mystery

🥇 Midjourney (left) 🥉 Adobe (center) 🥉 Meta (right). Similar to the previous round, Adobe Firefly produced a realistic image, but it did not accurately reflect the specifics of the prompt. Meta's output, on the other hand, fell short in terms of rendering quality and overall detail.

Links to images: Midjourney Adobe Firefly Meta AI

prompt: an old but rich filipino man wearing a beige suit, photo-realistic, leica m6, 50mm f/1.8 lens

🥇 Midjourney (left) 🥉 Adobe (center) 🥈 Meta (right). Adobe Firefly demonstrates its capability for generating high-quality, realistic images. However, this example highlights a potential limitation in the model's understanding of context, as the depicted individual does not appear to be old.

Links to images: Midjourney Adobe Firefly Meta AI

prompt: a man scoring a point in pickleball, sports photography, freezing the dynamic motion of victory on the pickleball court, with sharp focus and vibrant colors capturing the adrenaline-fueled triumph, slight motion blur

Links to images: Midjourney Adobe Firefly Meta AI

prompt: a young filipino-american woman wearing a cream sweater, in the style of mamiya rb67, shige's visual aesthetic style, dark brown and light beige, tumblewave, oshare kei, brooding mood

Links to images: Midjourney Adobe Firefly Meta AI

prompt: cinematic, high key photo, a curly long-haired man, ARRIFLEX 35 BL camera, canon k35 prime lenses, black and white, subtlety, model photography

Links to images: Midjourney Adobe Firefly Meta AI

prompt: commercial photography of a windbreaker with pastel 80s vaporwave aesthetic, woman, streetwear, fashion, mamiya rb67

🥇 Midjourney (left) 🥈 Adobe (center) 🥇 Meta (right). All the AI image generators successfully interpreted the prompt. However, the image produced by Adobe Firefly has some rendering issues.

Links to images: Midjourney Adobe Firefly Meta AI

prompt: 2020s fashion, full body of a 23-year-old japanese man sitting, minimalist and comfort fashion, snapshot

Links to images: Midjourney Adobe Firefly Meta AI

prompt: a woman lying in bed with her eyes closed, golden hour, closeup

🥇 Midjourney (left) 🥈 Adobe (center) 🥈 Meta (right). All AIs understood the prompt. Midjourney's image is stunningly detailed, especially the skin. Adobe's is detailed too, but it looks like a stock photo with the usual smiling person.

Links to images: Midjourney Adobe Firefly Meta AI

prompt: closeup portrait photo of a young dutch woman looking in the sky during winter

🥇 Midjourney (left) 🥈 Adobe (center) 🥇 Meta (right). Mijorney is the winner, but Meta generated a very aesthetic image this time, so we decided to rate them equally.

Links to images: Midjourney Adobe Firefly Meta AI

prompt: aesthetic photography, close up portrait of beautiful blonde woman with blue eyes, calm atmosphere, warm colors, snapshot photography, tapestry of beauty

Links to images: Midjourney Adobe Firefly Meta AI

prompt: high quality photography of a young Japanese woman smiling, backlighting, natural pale light, film camera, by Rinko Kawauchi, HDR

Links to images: Midjourney Adobe Firefly Meta AI

prompt: redhead boy wearing black hoodie with cyan flower pattern. Photograph by Loretta Lux

Links to images: Midjourney Adobe Firefly Meta AI

prompt: outdoor shoot highlighting bohemian style, a harmony of earthy tones, model adorned with floral headpiece, layers of textured fabrics, sun-kissed skin, amidst a wildflower field

Links to images: Midjourney Adobe Firefly Meta AI

Results and conclusion:

#1 🏆 Midjourney (🥇 x 25)

#2 🥈 Adobe (🥇 x 2, 🥈x 13, 🥉x 9)

#2 🥈 Meta (🥇 x 5, 🥈x 14, 🥉x 6)

Midjourney is the undisputed champion, consistently delivering exceptional results and earning a gold medal in every round. For professional purposes, such as creating logos, website imagery, illustrations and digital art, Midjourney's ability to produce highly detailed, visually stunning images makes it the top choice, despite its paid subscription model.

Adobe Firefly demonstrated the ability to generate sharp, detailed images, constantly surpassing Meta in this regard. However, in our testing, where a uniform prompt was used for all AIs, Adobe Firefly often failed to fully capture the intended context, leading to images that deviated from the prompt’s specifics. It's worth noting that Adobe Firefly's capabilities extend beyond those tested in our evaluation. Its integration with other Adobe products like Photoshop and InDesign, along with its advanced settings, offer greater control and flexibility, especially for users familiar with the Adobe ecosystem. With further development, Adobe Firefly has the potential to become a strong competitor in the AI image generation landscape.

Meta AI image generator presents a compelling option for users seeking a free, unlimited, and swift tool. While its rendering quality and detail may not reach the same level as its competitors, it remains a practical choice for rapidly generating conceptual and realistic images or for creative experimentation. However, Meta may fall short of expectations for professional use cases, and its persistent watermark is a notable drawback.

DALL-E 3, another prominent AI image generator, deserves an honorable mention, although it was not included in this particular review due to its limitations in generating realistic portraits. Its overall quality would likely rank it second, below Midjourney but above Adobe Firefly and Meta. A dedicated review exploring DALL-E's strengths and weaknesses will be released in the future.

The AI image generation field is rapidly evolving, with each platform offering its advantages and drawbacks. While Midjourney currently leads the pack, the competition remains intense, and advancements are constantly being made.

Bonus

We've curated a selection of different, stunning images generated by Midjourney that showcase the incredible capabilities of this AI image generator.

prompt: Audi Q8 50 TDI, 2023 car black realistic Download image

prompt: landscape, an autumn in the lake during dusk, tranquility Download image

prompt: a photo of Pikachu meeting the Queen of England Download image

prompt: a black 2019 Volkwagen Golf GTI with snow groomer cat tracks instead of wheels. Plowing through snow heading up mountain Download image

prompt: a photo of tiny Pikachu and huge Chewbacca on a couch. Chewbacca trying to take game controller from Pikachu. Dark room Download image

prompt: commercial photography, a perfume bottle, pastel blue background, dreamy, soft lighting, centered, flowers Downoad image

prompt: rows upon rows of lavender stretch to the horizon under a full moon, in the style of vincent van gogh Download image

prompt: wide-angle shot capturing the sprawling expanse of a metropolitan skyline at dusk, with lights beginning to illuminate the buildings Download image

prompt: solitary stone cabin in a vast alpine meadow, wildflowers in bloom, snow-capped peaks in the distance Download image

prompt: commercial photography, a handcrafted ceramic bowl, earth tones, soft lighting, plants Download image

prompt: interior shot of a bathroom, luxury high end, beige colors, penthouse suite, architecture digest photography, sophisticated style Download image

prompt: film still, white angle shot of an astronaut walking on the moon, sci-fi movie, by christopher nolan, celestial bodies Download image

Midjourney v6 vs Adobe Firefly 2 vs Meta AI

Results and conclusion:

Bonus

Report Page