Head to head
Esc

Models · slot A

to navigate to pick

Nano Banana Pro Google HiDream E1 HiDream AI

Settled by community votes across 4 shared challenges, with an AI judge weighing in on each.

Nano Banana Pro

28.2 arena score

#2 of 62 in Text-to-Image

Best Image Editing right now Top 2 in Text-to-Image
Skill signature · Text-to-Image

HiDream E1

0.7 arena score

#33 of 32 in Image Editing

Vote tally

Where the votes landed

Nano Banana Pro

0%

win rate

Ties

0%

HiDream E1

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 Editing
Edit instruction

“Give the person a full, thick head of natural hair with realistic texture, density, and a natural hairline. Preserve facial features and lighting.”

Before After
Nano Banana Pro
Before After
HiDream E1

AI Judge Analysis

Nano Banana Pro

  • + Excellent source preservation, keeping the background and clothing identical to the original.
  • + Realistic hair texture and natural-looking hairline integration.
  • + Flawless preservation of lighting and facial features.

HiDream E1

  • + Successfully applied the requested edit of thick hair.
  • Completely failed to preserve the source image's background, changing the environment to a desert with hills.
  • The hair texture looks artificial and overly stylized compared to the rest of the image.
  • Significantly altered the facial features and jacket texture, introducing artifacts.

Verdict: Nano Banana Pro followed the instructions perfectly, adding a realistic head of hair while keeping every other detail of the source image intact. HiDream E1 failed as an image editor by completely regenerating the background and altering the subject's face, resulting in a low-quality output with unnatural hair. Nano Banana Pro is the clear winner for its accuracy and preservation of the original content.

Over-the-top cartoon caricature

Editing
Edit 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
Nano Banana Pro
HiDream E1

AI Judge Analysis

Nano Banana Pro

  • + Excellent adherence to all prompts: TV news set, multiple dogs, and hockey gear.
  • + Strong character likeness to the source image within an illustrative caricature style.
  • + High level of detail with humorous touches like the 'Anchor Lady' mug and hockey-playing dogs.
  • The style shifts completely to a 2D digital illustration rather than a photo-based caricature.

HiDream E1

  • + Includes the subject, a dog, and hockey players in the background.
  • + Follows the humorous instruction with a very exaggerated facial expression.
  • Loses the facial likeness of the source image entirely, changing hair color and features.
  • Anatomical issues with the hands and the way the dog is held.
  • Fails to clearly represent the 'TV anchor' profession beyond a red blazer.

Verdict: Nano Banana Pro is the clear winner as it successfully incorporates every element requested—TV anchor set, dogs, and hockey—in a coherent and humorous composition while maintaining the subject's likeness. HiDream E1 fails to preserve the subject's identity, changing her hair from brown to blonde, and has several anatomical rendering errors.

Studio Ghibli Anime Style

Editing
Edit 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
Nano Banana Pro
HiDream E1

AI Judge Analysis

Nano Banana Pro

  • + Excellently captures the watercolor and hand-painted texture typical of Studio Ghibli backgrounds.
  • + Maintains the specific poses and clothing details, such as the plaid shirt, from the source image.
  • + Successfully incorporates dreamy, nostalgic lighting and Japanese-style architecture in the background.
  • The facial expressions are slightly more melancholic than the original meme's 'distracted' and 'outraged' vibe.

HiDream E1

  • + Applies a clean anime aesthetic with soft pastel colors as requested.
  • + Preserves the basic spatial relationship between the characters.
  • Fails to preserve the specific clothing colors and patterns by changing the red dress to tan and the plaid shirt to pink.
  • The art style is more generic modern web-comic than the requested Studio Ghibli style.
  • Loses the hand-painted texture and environmental detail seen in the source and the other model.

Verdict: Nano Banana Pro is the clear winner as it perfectly captures the Studio Ghibli aesthetic with its hand-painted watercolor textures and atmospheric lighting while strictly preserving the source image's clothing and pose details. In contrast, HiDream E1 produces a more generic vector-style illustration and fails to maintain source consistency by changing the colors of the characters' outfits.

Golden Hour Stroll

Image Editing
Edit instruction

“Add dynamic motion to this photo: make hair blow in the wind, add leaves flying, energetic and lively feel.”

Before After
Nano Banana Pro
Before After
HiDream E1

AI Judge Analysis

Nano Banana Pro

  • + Excellent preservation of the original person and dog details.
  • + Adds realistic motion-blurred hair and dynamic leaves.
  • + Maintains the lighting and texture of the source image perfectly.
  • Some leaves appear a bit small or fragmented.

HiDream E1

  • + Successfully adds wind-blown hair and leaves as requested.
  • + The leaves are large and clearly visible.
  • Significantly alters the woman's facial features and hair color.
  • The dog's face and fur texture have been noticeably changed.
  • The added leaves look like flat, artificial stickers rather than integrated elements.

Verdict: Nano Banana Pro is the clear winner as it successfully follows the editing instructions while perfectly preserving the identity of the woman and the dog from the source image. In contrast, HiDream E1 fails as an editing model by regenerating the faces and textures, resulting in a loss of the original subject's likeness.

Next steps

Explore each model