# Image Model Bake-off: seven AI image models on 18 real jobs

> We ran seven AI image models on 18 real jobs, from edits to characters and exact text. GPT Image 2.5 Flare won the edits; Nano Banana Pro won nothing.

Source: https://nodaro.ai/docs/research/image-model-bakeoff

We gave seven AI image models the same 18 jobs a Nodaro user runs every day: edits, characters from a reference photo, scenes, exact text and seven built-in presets. **GPT Image 2.5 Flare** was the best editor. **GPT Image 2** was the best at characters and exact text, and the slowest. **Nano Banana 2.1** won the scenes and the illustration. **Nano Banana Pro**, a popular choice for character work, won nothing.

- Models: 7
- Prompts: 18
- Images: 112
- Rounds: 3
- Images per prompt: 1
- Picks: blind

## What we found
**GPT Image 2.5 Flare is the most reliable editor.** 
It won two of the three edits and kept the pose, the framing and the product design on all three.
  

  
**GPT Image 2 leads on characters and exact text.** 
It made the best reference grid, kept both people right on the cast sheet and drew the cleanest infographic. It is also the slowest, with a median of 97 seconds.
  

  
**Nano Banana 2.1 wins scenes and illustration.** 
It won the rainy street, the watercolor and the labelled cast sheet. It changed the outfit when it put a person in a new scene.
  

  
**Nano Banana Pro won none of 18 prompts.** 
It broke the grid layout, dropped a name label, changed outfits and returned a nearly empty infographic: 8 weak marks in 11 prompts.
  

  
**GPT Image 2 won four of the seven presets.** 
The character board, the product shot, the claymation scene and the age progression. Nano Banana 2.1 won the expression sheet and the 4:5 studio portrait, and GPT Image 2.5 Sunburst won the 2:3 movie poster.
  

  
**Seedream 5 Flash is the best model for its price.** 
At 11 credits and about 45 seconds an image, it tied the earlier pick on 7 of 18 prompts. It beat none, and it mixed up the two people on the cast sheet.
  

## The scoreboard
Round 1 ran five models on 11 prompts. A pick is the best result of a prompt, and a weak mark is a result that missed the brief.

| Model | Picks | Weak marks | Median time | Credits per 2K image |
| --- | --- | --- | --- | --- |
| [GPT Image 2.5 Flare](https://nodaro.ai/docs/models/image/gpt-image-2-5-flare) | 4 | 1 | 64 s | 28 |
| [GPT Image 2](https://nodaro.ai/docs/models/image/gpt-image-2) | 3 | 2 | 97 s | 42 |
| [Nano Banana 2.1](https://nodaro.ai/docs/models/image/nano-banana-2-1) | 3 | 1 | 68 s | 22 |
| [GPT Image 2.5 Sunburst](https://nodaro.ai/docs/models/image/gpt-image-2-5-sunburst) | 1 | 1 | 59 s | 28 |
| [Nano Banana Pro](https://nodaro.ai/docs/models/image/nano-banana-pro) | 0 | 8 | 51 s | 50 |

In round 2, the seven presets, GPT Image 2 won four, Nano Banana 2.1 two and GPT Image 2.5 Sunburst one. Nano Banana Pro won none again. In round 3, neither Seedream model beat an earlier pick.

## How we ran it
- Every model got the same prompt, aspect ratio and output size, on Nodaro, on 2026-10-08. Each model made one image per prompt.
- Round 1 ran five models on 11 prompts in four groups: edits, identity from a reference photo, text to image, and exact text.
- Round 2 ran seven of Nodaro's built-in presets on three models.
- Round 3 ran all 18 prompts again on Seedream 5 Pro and Seedream 5 Flash, and compared each result with the earlier pick.
- Claude, an AI model, made every pick and weak mark blind: it saw the results without the model names.
- One image per prompt is a hint, not proof. The pattern across prompts is the finding.
- The edits and the identity prompts start from four photos from Unsplash, shared on Wikimedia Commons under CC0: [Brunette woman portrait](https://commons.wikimedia.org/wiki/File:Brunette_woman_portrait_(Unsplash).jpg), [Gray-haired man portrait](https://commons.wikimedia.org/wiki/File:Gray-haired_man_portrait_(Unsplash).jpg), [Grado Headphones SR80e](https://commons.wikimedia.org/wiki/File:Grado_Headphones_SR80e_(Unsplash).jpg) and [Confident Eye Contact](https://commons.wikimedia.org/wiki/File:Confident_Eye_Contact_(Unsplash).jpg).

## Round 1: five models, 11 prompts
Each model made one image per prompt, at the same aspect ratio and size. The pick of each prompt is marked, and so is every weak result. The galleries show the models in the same order every time.

### Edits
Each edit starts from a photo and asks for one change. The rest of the photo must stay as it is.

#### Swap the clothes
```
Change her black top to a tailored emerald-green blazer over a plain white t-shirt. Keep her face, hair, pose, framing, background and lighting exactly the same.
```

  
![The starting photo: a young woman with long auburn hair in a black top, in front of a green hedge](https://nodaro.ai/docs-media/research/image-model-bakeoff/source/src-editsrc.webp)

The starting photo

  - **[GPT Image 2.5 Flare](https://nodaro.ai/docs/models/image/gpt-image-2-5-flare)** (Pick) · 50 s · 2368×1776 ![GPT Image 2.5 Flare's result for “Swap the clothes”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/edit-clothes-D.webp) Full-size original: https://cdn.nodaro.ai/images/90fe31a1-a581-47d0-84ff-97c3fa7059d5.png
  - **[GPT Image 2](https://nodaro.ai/docs/models/image/gpt-image-2)** · 97 s · 2688×2016 ![GPT Image 2's result for “Swap the clothes”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/edit-clothes-E.webp) Full-size original: https://cdn.nodaro.ai/images/eb00a1f4-dc58-42fd-94ca-98628c4de8ec.png
  - **[Nano Banana 2.1](https://nodaro.ai/docs/models/image/nano-banana-2-1)** · 68 s · 2400×1792 ![Nano Banana 2.1's result for “Swap the clothes”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/edit-clothes-C.webp) Full-size original: https://cdn.nodaro.ai/images/16ce4376-0c3f-40ac-b3f5-46e0095bcea3.png
  - **[GPT Image 2.5 Sunburst](https://nodaro.ai/docs/models/image/gpt-image-2-5-sunburst)** · 68 s · 2368×1776 ![GPT Image 2.5 Sunburst's result for “Swap the clothes”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/edit-clothes-A.webp) Full-size original: https://cdn.nodaro.ai/images/9c43d144-2c38-4e55-a80b-0e6df131a89c.png
  - **[Nano Banana Pro](https://nodaro.ai/docs/models/image/nano-banana-pro)** (Weak) · 51 s · 2400×1792. Moved the camera closer; framing drifted ![Nano Banana Pro's result for “Swap the clothes”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/edit-clothes-B.webp) Full-size original: https://cdn.nodaro.ai/images/1a8c1851-c0f6-43fa-8987-8ef8faf50688.png

#### Replace the background
```
Replace the grey studio background with a bright modern open-plan office, softly out of focus. Keep the man, his face, glasses, polo shirt, pose and lighting unchanged.
```

  
![The starting photo: a gray-haired man with glasses in a polo shirt, in front of a gray studio background](https://nodaro.ai/docs-media/research/image-model-bakeoff/source/src-personB.webp)

The starting photo

  - **[GPT Image 2.5 Flare](https://nodaro.ai/docs/models/image/gpt-image-2-5-flare)** · 126 s · 1776×2368. Kept pose and framing exactly ![GPT Image 2.5 Flare's result for “Replace the background”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/edit-background-D.webp) Full-size original: https://cdn.nodaro.ai/images/14c57157-5e36-4cf6-940c-172e434bb1c3.png
  - **[GPT Image 2](https://nodaro.ai/docs/models/image/gpt-image-2)** (Weak) · 97 s · 2016×2688. Changed his pose and crop ![GPT Image 2's result for “Replace the background”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/edit-background-E.webp) Full-size original: https://cdn.nodaro.ai/images/ae6b1f2a-bdec-42b5-8301-dc31f5547e5a.png
  - **[Nano Banana 2.1](https://nodaro.ai/docs/models/image/nano-banana-2-1)** · 97 s · 1792×2400 ![Nano Banana 2.1's result for “Replace the background”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/edit-background-C.webp) Full-size original: https://cdn.nodaro.ai/images/85db7abc-612c-49f4-b9f9-17a1487bc724.png
  - **[GPT Image 2.5 Sunburst](https://nodaro.ai/docs/models/image/gpt-image-2-5-sunburst)** (Pick) · 56 s · 1776×2368 ![GPT Image 2.5 Sunburst's result for “Replace the background”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/edit-background-A.webp) Full-size original: https://cdn.nodaro.ai/images/a0b42c94-a29c-4f79-9480-9294b7a6c03c.png
  - **[Nano Banana Pro](https://nodaro.ai/docs/models/image/nano-banana-pro)** (Weak) · 44 s · 1792×2400. Shrank and moved him in frame ![Nano Banana Pro's result for “Replace the background”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/edit-background-B.webp) Full-size original: https://cdn.nodaro.ai/images/8b55b66d-2983-4b29-b509-1a893f8ed59f.png

#### Restage a product photo
```
Remove the hand. Place these exact headphones on a walnut desk next to a closed laptop, in warm morning window light, premium product photography. Keep the headphones' exact design, materials, colours and logo.
```

  
![The starting photo: a hand holding a pair of headphones](https://nodaro.ai/docs-media/research/image-model-bakeoff/source/src-product.webp)

The starting photo

  - **[GPT Image 2.5 Flare](https://nodaro.ai/docs/models/image/gpt-image-2-5-flare)** (Pick) · 86 s · 2368×1776 ![GPT Image 2.5 Flare's result for “Restage a product photo”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/edit-product-A.webp) Full-size original: https://cdn.nodaro.ai/images/d6720ab3-92ec-4a69-bf5f-cdc27f3c6bdf.png
  - **[GPT Image 2](https://nodaro.ai/docs/models/image/gpt-image-2)** · 127 s · 2688×2016 ![GPT Image 2's result for “Restage a product photo”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/edit-product-C.webp) Full-size original: https://cdn.nodaro.ai/images/47c591c7-599a-47e8-96b6-3cd4cbf559c4.png
  - **[Nano Banana 2.1](https://nodaro.ai/docs/models/image/nano-banana-2-1)** · 68 s · 2400×1792. Correct but flatter light ![Nano Banana 2.1's result for “Restage a product photo”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/edit-product-B.webp) Full-size original: https://cdn.nodaro.ai/images/1f20ed2a-d5c5-4920-b57d-8e9ea8b6aaf6.png
  - **[GPT Image 2.5 Sunburst](https://nodaro.ai/docs/models/image/gpt-image-2-5-sunburst)** · 50 s · 2368×1776 ![GPT Image 2.5 Sunburst's result for “Restage a product photo”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/edit-product-D.webp) Full-size original: https://cdn.nodaro.ai/images/2ee77553-b948-49c9-b29b-fc88a5319291.png
  - **[Nano Banana Pro](https://nodaro.ai/docs/models/image/nano-banana-pro)** (Weak) · 51 s · 2400×1792. Shrank the headphones and merged the cups ![Nano Banana Pro's result for “Restage a product photo”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/edit-product-E.webp) Full-size original: https://cdn.nodaro.ai/images/38c4bc10-dedf-4c51-a01a-282d4e639e74.png

### Identity from a reference photo
Each prompt starts from a photo of a person, and the person must stay recognizable.

#### Put the person in a new scene
```
The woman from the reference photo sitting at a small café table on a Paris street, laughing, holding a cup of coffee, natural daylight, candid 35mm photograph. Keep her exact face, skin tone and hair.
```

  
![The starting photo: a young woman looking straight at the camera](https://nodaro.ai/docs-media/research/image-model-bakeoff/source/src-personA.webp)

The starting photo

  - **[GPT Image 2.5 Flare](https://nodaro.ai/docs/models/image/gpt-image-2-5-flare)** (Pick) · 51 s · 1776×2368 ![GPT Image 2.5 Flare's result for “Put the person in a new scene”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/id-scene-C.webp) Full-size original: https://cdn.nodaro.ai/images/fa2b85c2-c609-474a-bbde-033d951bbcd0.png
  - **[GPT Image 2](https://nodaro.ai/docs/models/image/gpt-image-2)** · 79 s · 2016×2688. Face, hair and top kept ![GPT Image 2's result for “Put the person in a new scene”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/id-scene-A.webp) Full-size original: https://cdn.nodaro.ai/images/88d98abb-56de-4cf4-bf15-48ce00b2afaf.png
  - **[Nano Banana 2.1](https://nodaro.ai/docs/models/image/nano-banana-2-1)** (Weak) · 68 s · 1792×2400. Changed her top to a black coat; face small in frame ![Nano Banana 2.1's result for “Put the person in a new scene”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/id-scene-B.webp) Full-size original: https://cdn.nodaro.ai/images/38d01c97-d1da-47b0-94d4-c43da4f6429a.png
  - **[GPT Image 2.5 Sunburst](https://nodaro.ai/docs/models/image/gpt-image-2-5-sunburst)** · 56 s · 1776×2368 ![GPT Image 2.5 Sunburst's result for “Put the person in a new scene”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/id-scene-E.webp) Full-size original: https://cdn.nodaro.ai/images/52a3be96-5175-4da2-9d76-a6f513cacc99.png
  - **[Nano Banana Pro](https://nodaro.ai/docs/models/image/nano-banana-pro)** (Weak) · 51 s · 1792×2400. Changed her top ![Nano Banana Pro's result for “Put the person in a new scene”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/id-scene-D.webp) Full-size original: https://cdn.nodaro.ai/images/6d916018-4407-4b2a-824f-dd41b29bcb1e.png

#### Character reference grid
```
Create a clean character reference grid using the attached photo of the person as the single source of truth for face, hair, eyes, skin tone, build and outfit. Four equal panels arranged in a tidy two-by-two grid on a single seamless neutral light-grey studio background: a tight front-facing portrait close-up, a 3/4-view portrait, a full-body shot from the front in a relaxed neutral stance, and a full-body shot from the back. The SAME person in every panel — identical age, features, hairstyle and outfit from the attached photo. Identical soft, even studio lighting in all four panels with no dramatic shadows. No decorations, no background props, no text, no UI elements — a sterile, production-neutral reference sheet built to be reused as an identity reference in later generations. Photorealistic, sharp focus, accurate natural skin texture, true-to-life color.
```

  
![The starting photo: a young woman looking straight at the camera](https://nodaro.ai/docs-media/research/image-model-bakeoff/source/src-personA.webp)

The starting photo

  - **[GPT Image 2.5 Flare](https://nodaro.ai/docs/models/image/gpt-image-2-5-flare)** · 68 s · 2480×3312 ![GPT Image 2.5 Flare's result for “Character reference grid”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/id-grid-B.webp) Full-size original: https://cdn.nodaro.ai/images/5a9e91f7-d394-4dd4-983e-1c1610d333bf.png
  - **[GPT Image 2](https://nodaro.ai/docs/models/image/gpt-image-2)** (Pick) · 79 s · 2448×3264 ![GPT Image 2's result for “Character reference grid”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/id-grid-A.webp) Full-size original: https://cdn.nodaro.ai/images/7d73dded-44c2-472c-a4c2-ef3366e78a8e.png
  - **[Nano Banana 2.1](https://nodaro.ai/docs/models/image/nano-banana-2-1)** · 121 s · 3584×4800 ![Nano Banana 2.1's result for “Character reference grid”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/id-grid-E.webp) Full-size original: https://cdn.nodaro.ai/images/252d8e2d-5120-49dd-b3c1-ccf7601b8088.png
  - **[GPT Image 2.5 Sunburst](https://nodaro.ai/docs/models/image/gpt-image-2-5-sunburst)** · 98 s · 2480×3312 ![GPT Image 2.5 Sunburst's result for “Character reference grid”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/id-grid-D.webp) Full-size original: https://cdn.nodaro.ai/images/815c6289-d0f5-4f89-9c4b-80be4b09e811.png
  - **[Nano Banana Pro](https://nodaro.ai/docs/models/image/nano-banana-pro)** (Weak) · 68 s · 3584×4800. Broke the layout: no full-body front, cropped back view ![Nano Banana Pro's result for “Character reference grid”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/id-grid-C.webp) Full-size original: https://cdn.nodaro.ai/images/54a5c46f-169e-4c9b-8527-e95505eb6449.png

#### Two-person cast sheet
```
Create a clean multi-character cast reference sheet composed as horizontal strips stacked vertically in one image — one strip per character from the attached reference images, in the order they are attached, using exactly as many strips as there are attached characters. Each strip shows that character in four aligned studio panels: tight front portrait close-up · 3/4 portrait · full body front · full body back, all in a relaxed neutral stance. Label each strip on the left edge with the character's name in clean bold lettering: MAYA · DAVID. Every character must match their attached reference exactly — same face, hair, skin tone, build and outfit — with no identity blending between rows. Single seamless neutral light-grey studio background across the whole sheet, identical soft even lighting in every panel, consistent panel sizing and alignment. No decorations, no props, no UI elements beyond the row name labels. A sterile production casting sheet built to be reused as the single identity reference for this cast in later generations. Photorealistic, sharp focus, natural skin texture, true-to-life color.
```

  
![The starting photo: a young woman looking straight at the camera](https://nodaro.ai/docs-media/research/image-model-bakeoff/source/src-personA.webp)

The starting photo

  
![The starting photo: a gray-haired man with glasses in a polo shirt, in front of a gray studio background](https://nodaro.ai/docs-media/research/image-model-bakeoff/source/src-personB.webp)

The starting photo

  - **[GPT Image 2.5 Flare](https://nodaro.ai/docs/models/image/gpt-image-2-5-flare)** · 69 s · 2480×3312 ![GPT Image 2.5 Flare's result for “Two-person cast sheet”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/id-cast-B.webp) Full-size original: https://cdn.nodaro.ai/images/467449c1-b778-4283-abda-95d0d2f3671e.png
  - **[GPT Image 2](https://nodaro.ai/docs/models/image/gpt-image-2)** · 109 s · 2448×3264. Both identities right; labels placed oddly ![GPT Image 2's result for “Two-person cast sheet”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/id-cast-D.webp) Full-size original: https://cdn.nodaro.ai/images/d10d2975-3b8e-4459-8219-29c11577d506.png
  - **[Nano Banana 2.1](https://nodaro.ai/docs/models/image/nano-banana-2-1)** (Pick) · 98 s · 3584×4800 ![Nano Banana 2.1's result for “Two-person cast sheet”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/id-cast-A.webp) Full-size original: https://cdn.nodaro.ai/images/969d9f79-7684-4a89-958f-364d5494b12c.png
  - **[GPT Image 2.5 Sunburst](https://nodaro.ai/docs/models/image/gpt-image-2-5-sunburst)** (Weak) · 85 s · 2480×3312. Dropped the MAYA label ![GPT Image 2.5 Sunburst's result for “Two-person cast sheet”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/id-cast-E.webp) Full-size original: https://cdn.nodaro.ai/images/2be87fcd-930d-4435-8550-a1829b9309ec.png
  - **[Nano Banana Pro](https://nodaro.ai/docs/models/image/nano-banana-pro)** (Weak) · 68 s · 3584×4800. Dropped the MAYA label and changed David's shirt ![Nano Banana Pro's result for “Two-person cast sheet”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/id-cast-C.webp) Full-size original: https://cdn.nodaro.ai/images/23be04a7-0d44-4246-9842-4267c92fbf0e.png

### Text to image
Three prompts with no starting image: a scene, a product and an illustration.

#### Rainy street at night
```
A rainy Tokyo side street at night, neon signs reflecting on wet asphalt, a lone cyclist with a clear umbrella passing a ramen shop, cinematic photograph, shallow depth of field.
```

  - **[GPT Image 2.5 Flare](https://nodaro.ai/docs/models/image/gpt-image-2-5-flare)** (Weak) · 56 s · 2048×1152. Lost the clear umbrella ![GPT Image 2.5 Flare's result for “Rainy street at night”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/t2i-street-C.webp) Full-size original: https://cdn.nodaro.ai/images/1cc3199c-8cbb-4a92-b8a9-5a571b51f482.png
  - **[GPT Image 2](https://nodaro.ai/docs/models/image/gpt-image-2)** (Weak) · 109 s · 2048×1152. Lost the clear umbrella ![GPT Image 2's result for “Rainy street at night”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/t2i-street-E.webp) Full-size original: https://cdn.nodaro.ai/images/c95223ff-81aa-4ebf-a80f-e12f7ba6320a.png
  - **[Nano Banana 2.1](https://nodaro.ai/docs/models/image/nano-banana-2-1)** (Pick) · 97 s · 2752×1536 ![Nano Banana 2.1's result for “Rainy street at night”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/t2i-street-B.webp) Full-size original: https://cdn.nodaro.ai/images/811306c9-9a46-426b-9f29-c70e956a9276.png
  - **[GPT Image 2.5 Sunburst](https://nodaro.ai/docs/models/image/gpt-image-2-5-sunburst)** · 68 s · 2736×1536 ![GPT Image 2.5 Sunburst's result for “Rainy street at night”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/t2i-street-A.webp) Full-size original: https://cdn.nodaro.ai/images/af8905e2-49e8-4346-b9cc-67eb12321a5f.png
  - **[Nano Banana Pro](https://nodaro.ai/docs/models/image/nano-banana-pro)** · 44 s · 2752×1536 ![Nano Banana Pro's result for “Rainy street at night”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/t2i-street-D.webp) Full-size original: https://cdn.nodaro.ai/images/7275c2c9-484a-420b-923a-7f7bc255a4fa.png

#### Product photo of a mug
```
Studio product photograph of a matte black ceramic coffee mug with a wisp of steam, on a light oak surface, soft diffused window light, minimal clean background.
```

  - **[GPT Image 2.5 Flare](https://nodaro.ai/docs/models/image/gpt-image-2-5-flare)** · 56 s · 2048×2048 ![GPT Image 2.5 Flare's result for “Product photo of a mug”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/t2i-product-C.webp) Full-size original: https://cdn.nodaro.ai/images/306ba90f-de98-410d-aa8a-c8c9c8e05149.png
  - **[GPT Image 2](https://nodaro.ai/docs/models/image/gpt-image-2)** (Pick) · 68 s · 2048×2048 ![GPT Image 2's result for “Product photo of a mug”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/t2i-product-A.webp) Full-size original: https://cdn.nodaro.ai/images/eec22388-4123-489d-92e2-4b69cb005dce.png
  - **[Nano Banana 2.1](https://nodaro.ai/docs/models/image/nano-banana-2-1)** · 68 s · 2048×2048. Usable; odd handle shape ![Nano Banana 2.1's result for “Product photo of a mug”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/t2i-product-D.webp) Full-size original: https://cdn.nodaro.ai/images/c028ba84-7258-4877-9170-a1c65951b458.png
  - **[GPT Image 2.5 Sunburst](https://nodaro.ai/docs/models/image/gpt-image-2-5-sunburst)** · 57 s · 2048×2048 ![GPT Image 2.5 Sunburst's result for “Product photo of a mug”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/t2i-product-B.webp) Full-size original: https://cdn.nodaro.ai/images/f11b86fd-3e0a-4f99-87a9-faa50e74ebac.png
  - **[Nano Banana Pro](https://nodaro.ai/docs/models/image/nano-banana-pro)** · 45 s · 2048×2048 ![Nano Banana Pro's result for “Product photo of a mug”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/t2i-product-E.webp) Full-size original: https://cdn.nodaro.ai/images/57b37183-bedd-4170-8e76-b2cebc854430.png

#### Watercolor picture-book illustration
```
Children's picture-book watercolor illustration of a fox and a hedgehog sharing a red umbrella in an autumn forest, soft washes, warm palette, gentle linework.
```

  - **[GPT Image 2.5 Flare](https://nodaro.ai/docs/models/image/gpt-image-2-5-flare)** · 68 s · 2368×1776 ![GPT Image 2.5 Flare's result for “Watercolor picture-book illustration”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/t2i-illustration-A.webp) Full-size original: https://cdn.nodaro.ai/images/78596fc7-56af-4df5-af3f-70fd48cf05b9.png
  - **[GPT Image 2](https://nodaro.ai/docs/models/image/gpt-image-2)** · 142 s · 2688×2016 ![GPT Image 2's result for “Watercolor picture-book illustration”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/t2i-illustration-B.webp) Full-size original: https://cdn.nodaro.ai/images/5690180d-88e2-4a19-b1c9-a72aedf9b503.png
  - **[Nano Banana 2.1](https://nodaro.ai/docs/models/image/nano-banana-2-1)** (Pick) · 59 s · 2400×1792 ![Nano Banana 2.1's result for “Watercolor picture-book illustration”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/t2i-illustration-D.webp) Full-size original: https://cdn.nodaro.ai/images/c0a5aeaf-7af0-46ea-8f3c-e77b3a079d57.png
  - **[GPT Image 2.5 Sunburst](https://nodaro.ai/docs/models/image/gpt-image-2-5-sunburst)** · 50 s · 2368×1776 ![GPT Image 2.5 Sunburst's result for “Watercolor picture-book illustration”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/t2i-illustration-E.webp) Full-size original: https://cdn.nodaro.ai/images/c831f499-00da-4889-a5d2-bcc2ffdf6772.png
  - **[Nano Banana Pro](https://nodaro.ai/docs/models/image/nano-banana-pro)** · 50 s · 2400×1792. Close second; taste call ![Nano Banana Pro's result for “Watercolor picture-book illustration”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/t2i-illustration-C.webp) Full-size original: https://cdn.nodaro.ai/images/8f54b134-2335-4f4f-9fcf-c0c54e97bfe3.png

### Exact text
Two prompts that spell out every word the image must show.

#### Concert poster with exact text
```
Concert poster with this exact text, spelled exactly: headline "MIDNIGHT ECHOES", subline "Live at the Blue Room", details "Friday 14 November · 9 PM", footer "Tickets $25". Bold modern typography, deep navy and gold, clean layout.
```

  - **[GPT Image 2.5 Flare](https://nodaro.ai/docs/models/image/gpt-image-2-5-flare)** (Pick) · 59 s · 1776×2368 ![GPT Image 2.5 Flare's result for “Concert poster with exact text”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/type-poster-B.webp) Full-size original: https://cdn.nodaro.ai/images/2d6dfa69-87c3-47f0-8322-a8f17f35753a.png
  - **[GPT Image 2](https://nodaro.ai/docs/models/image/gpt-image-2)** · 85 s · 2016×2688 ![GPT Image 2's result for “Concert poster with exact text”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/type-poster-A.webp) Full-size original: https://cdn.nodaro.ai/images/9dbd688c-9df8-47ce-af83-b56362bbfe32.png
  - **[Nano Banana 2.1](https://nodaro.ai/docs/models/image/nano-banana-2-1)** · 53 s · 1792×2400 ![Nano Banana 2.1's result for “Concert poster with exact text”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/type-poster-C.webp) Full-size original: https://cdn.nodaro.ai/images/8066382a-136b-45d6-b51a-ecb034a2329e.png
  - **[GPT Image 2.5 Sunburst](https://nodaro.ai/docs/models/image/gpt-image-2-5-sunburst)** · 60 s · 1776×2368 ![GPT Image 2.5 Sunburst's result for “Concert poster with exact text”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/type-poster-D.webp) Full-size original: https://cdn.nodaro.ai/images/ada12cfa-2b8e-48b4-ae2f-fb0a5b352511.png
  - **[Nano Banana Pro](https://nodaro.ai/docs/models/image/nano-banana-pro)** (Weak) · 43 s · 1792×2400. Rendered as a skewed photo of a print ![Nano Banana Pro's result for “Concert poster with exact text”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/type-poster-E.webp) Full-size original: https://cdn.nodaro.ai/images/1903949a-514f-4849-91a0-17bd665d39a8.png

#### Infographic with exact labels
```
Clean flat infographic titled "How Coffee Gets to Your Cup" with four numbered steps, each with a simple icon and one short label: 1 Harvest, 2 Dry, 3 Roast, 4 Brew. Exact spelling, white background, friendly brown palette.
```

  - **[GPT Image 2.5 Flare](https://nodaro.ai/docs/models/image/gpt-image-2-5-flare)** · 1776×2368 ![GPT Image 2.5 Flare's result for “Infographic with exact labels”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/type-infographic-A.webp) Full-size original: https://cdn.nodaro.ai/images/44f7a084-bba9-4009-98f0-4a47ade59149.png
  - **[GPT Image 2](https://nodaro.ai/docs/models/image/gpt-image-2)** (Pick) · 80 s · 2016×2688 ![GPT Image 2's result for “Infographic with exact labels”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/type-infographic-D.webp) Full-size original: https://cdn.nodaro.ai/images/a5cd4e9e-11de-4e8a-982e-ed531c2c2e37.png
  - **[Nano Banana 2.1](https://nodaro.ai/docs/models/image/nano-banana-2-1)** · 1792×2400 ![Nano Banana 2.1's result for “Infographic with exact labels”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/type-infographic-C.webp) Full-size original: https://cdn.nodaro.ai/images/7a4dd6c1-381a-4389-999b-3934870b4319.png
  - **[GPT Image 2.5 Sunburst](https://nodaro.ai/docs/models/image/gpt-image-2-5-sunburst)** · 1776×2368 ![GPT Image 2.5 Sunburst's result for “Infographic with exact labels”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/type-infographic-B.webp) Full-size original: https://cdn.nodaro.ai/images/37f4899f-2478-4e6f-845d-9ee3ae5c2efe.png
  - **[Nano Banana Pro](https://nodaro.ai/docs/models/image/nano-banana-pro)** (Weak) · 1792×2400. Tiny and mostly empty ![Nano Banana Pro's result for “Infographic with exact labels”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/type-infographic-E.webp) Full-size original: https://cdn.nodaro.ai/images/082e0849-ad28-4cd3-8eaf-3202bf478b1a.png

## Round 2: seven built-in presets
Nodaro's presets are ready-made prompts for one job each. We ran seven of them on Nano Banana Pro, Nano Banana 2.1 and GPT Image 2, each with its own prompt and ratio. The Movie Poster preset is 2:3, a ratio GPT Image 2 cannot draw, so it ran on GPT Image 2.5 Sunburst instead.

### Character Board
```
Create a single high-resolution, densely packed character reference sheet titled "CHARACTER BOARD" using the attached photo of the person as the single source of truth for face, hair, beard, eyes, skin tone and body proportions. The same person must appear in every panel — same age, same features. Outfit identical to the attached photo across all panels. All on-image labels in ENGLISH. Editorial reference-board layout with a dark near-black background, thin yellow neon accent light, faint film-grain overlay, and production-grade character-design UI. The composition should feel organized but not rigidly locked: allow the main portrait, metadata, and supporting panels to shift position naturally within the board. The panel arrangement can vary from generation to generation while remaining readable, balanced, and premium. Design the board so it works cleanly in different aspect ratios without depending on a fixed left-column structure. Include: A large hero portrait or 3/4-length character image, accompanied by a detailed metadata block: NAME · AGE · HEIGHT · BUILD · HAIR · EYES · FEATURES · OUTFIT (describe the exact clothing from the attached photo) · CHARACTER · MOOD. Also include six content groupings — render ALL six panel headings exactly as written, never merging or omitting a panel: PANEL 01 — VIEWS (5 full-body shots, neutral pose, identical outfit and lighting, neutral tan backdrop): FRONT · 3/4 LEFT · SIDE LEFT · BACK · 3/4 RIGHT. PANEL 02 — EXPRESSIONS (5 tight headshots, same lighting): CALM · LAUGHING · INTENSE · CONTEMPLATIVE · CONFIDENT. PANEL 03 — DETAILS (3 macros): face/eyes close-up · hand close-up (showing ring, watch or cuff) · distinctive outfit detail. PANEL 04 — OUTFIT FLAT-LAYS (5 isolated product shots on dark background): outerwear · top · bottom · footwear · main accessory. PANEL 05 — LIGHTING / MOOD (4 same-pose portraits under different lighting): SOFT WINDOW LIGHT · GOLDEN HOUR · COOL BLUE NIGHT · DRAMATIC RIM-LIGHT. PANEL 06 — COLOR PALETTE: 6 swatches with HEX codes derived from the outfit and skin tones. Bottom caption: "Use this character board as a visual reference for consistent depiction of the character across all generations." Bottom-right tags: STYLE · Modern · Realistic · Cinematic. Style: photorealistic, no illustration. Consistent face across every single panel. 8K, fine grain, cinematic color grading.
```

  
![The starting photo: a young woman looking straight at the camera](https://nodaro.ai/docs-media/research/image-model-bakeoff/source/src-personA.webp)

The starting photo

  - **[GPT Image 2](https://nodaro.ai/docs/models/image/gpt-image-2)** (Pick) · 111 s · 2048×1152. Densest board, every panel, closest to the photo ![GPT Image 2's result for “Character Board”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r2/character-board-gpt-image-2.webp) Full-size original: https://cdn.nodaro.ai/images/fd06a91f-1baf-401a-8fa8-d3fdc3c22b06.png
  - **[Nano Banana 2.1](https://nodaro.ai/docs/models/image/nano-banana-2-1)** · 98 s · 2752×1536 ![Nano Banana 2.1's result for “Character Board”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r2/character-board-nano-banana-2-1.webp) Full-size original: https://cdn.nodaro.ai/images/045c704f-c86f-4d52-873f-be5b703b51d1.png
  - **[Nano Banana Pro](https://nodaro.ai/docs/models/image/nano-banana-pro)** · 56 s · 2752×1536. Invented a name; layout fine ![Nano Banana Pro's result for “Character Board”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r2/character-board-nano-banana-pro.webp) Full-size original: https://cdn.nodaro.ai/images/f8f6fdc0-14c3-469f-adaf-89142eeabd05.png

### Character Sheet · Expressions ×6
```
character expression model sheet of a cheerful freckled teen with messy red hair, six head-and-shoulders portraits arranged in two even rows of three, labeled emotions in order — neutral, happy, angry, surprised, sad, determined — strictly consistent face, hairstyle, and features across every panel with only the expression changing, flat even portrait lighting identical in all six panels, evenly spaced on a plain neutral backdrop, clean concept-art model sheet, natural facial detail and matching color palette
```

  - **[GPT Image 2](https://nodaro.ai/docs/models/image/gpt-image-2)** (Weak) · 80 s · 1448×1086. Photographic, but dropped the emotion labels ![GPT Image 2's result for “Character Sheet · Expressions ×6”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r2/character-expressions-6-gpt-image-2.webp) Full-size original: https://cdn.nodaro.ai/images/c5f9bd3e-de7a-4076-98b7-dd7908ae5425.png
  - **[Nano Banana 2.1](https://nodaro.ai/docs/models/image/nano-banana-2-1)** (Pick) · 49 s · 1200×896. Labelled, one consistent face ![Nano Banana 2.1's result for “Character Sheet · Expressions ×6”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r2/character-expressions-6-nano-banana-2-1.webp) Full-size original: https://cdn.nodaro.ai/images/4b27dd4f-a785-4e11-9c0b-78524ca1bd55.png
  - **[Nano Banana Pro](https://nodaro.ai/docs/models/image/nano-banana-pro)** · 36 s · 1200×896 ![Nano Banana Pro's result for “Character Sheet · Expressions ×6”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r2/character-expressions-6-nano-banana-pro.webp) Full-size original: https://cdn.nodaro.ai/images/1f6c1985-0f94-4f12-ba7e-eacb8d2d2d9a.png

### Product Shot (white bg)
```
e-commerce studio product photograph of a matte-black wireless headphone, on a seamless pure-white background, 100mm lens at f/8 with deep focus, soft even high-key lighting from a large diffused softbox with gentle fill, perfectly centered with generous margins, clean catalog-ready mood with true-to-life color, crisp edges and accurate material texture
```

  - **[GPT Image 2](https://nodaro.ai/docs/models/image/gpt-image-2)** (Pick) · 65 s · 1024×1024 ![GPT Image 2's result for “Product Shot (white bg)”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r2/product-shot-gpt-image-2.webp) Full-size original: https://cdn.nodaro.ai/images/afe4a2c5-7972-4434-b873-c5dbdc81e87c.png
  - **[Nano Banana 2.1](https://nodaro.ai/docs/models/image/nano-banana-2-1)** · 41 s · 1024×1024 ![Nano Banana 2.1's result for “Product Shot (white bg)”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r2/product-shot-nano-banana-2-1.webp) Full-size original: https://cdn.nodaro.ai/images/7ae7cec9-928c-41a0-b919-d10309858ee4.png
  - **[Nano Banana Pro](https://nodaro.ai/docs/models/image/nano-banana-pro)** (Weak) · 40 s · 1024×1024. Product small; a softbox in frame ![Nano Banana Pro's result for “Product Shot (white bg)”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r2/product-shot-nano-banana-pro.webp) Full-size original: https://cdn.nodaro.ai/images/981fbb57-6799-46d1-b622-bc9720a3b803.png

### Movie Poster
```
theatrical movie poster for 'NEON TIDE', cinematic key art in a gritty sci-fi thriller mood, dramatic vertical hero composition with a commanding central figure, moody atmospheric lighting with strong rim light and volumetric haze, rich filmic teal-and-amber color grade, bold title treatment along the bottom with a short tagline above it and a small billing-block strip, premium high-detail finish, legible correctly-spelled text
```

  - **[Nano Banana 2.1](https://nodaro.ai/docs/models/image/nano-banana-2-1)** · 27 s · 848×1264. Garbled small credits at the top ![Nano Banana 2.1's result for “Movie Poster”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r2/movie-poster-nano-banana-2-1.webp) Full-size original: https://cdn.nodaro.ai/images/3ab59745-6f00-4180-a181-84f8364e65dc.png
  - **[GPT Image 2.5 Sunburst](https://nodaro.ai/docs/models/image/gpt-image-2-5-sunburst)** (Pick) · 94 s · 1024×1536 ![GPT Image 2.5 Sunburst's result for “Movie Poster”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r2/movie-poster-gpt-image-2-5-sunburst.webp) Full-size original: https://cdn.nodaro.ai/images/08d21907-b434-4fa5-af60-a4a029aabc52.png
  - **[Nano Banana Pro](https://nodaro.ai/docs/models/image/nano-banana-pro)** · 34 s · 848×1264 ![Nano Banana Pro's result for “Movie Poster”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r2/movie-poster-nano-banana-pro.webp) Full-size original: https://cdn.nodaro.ai/images/45ce6336-d633-4a96-aa41-bba2fd9293fb.png

### Claymation Scene
```
a tiny clay fisherman rowing across a stormy plasticine sea, tactile stop-motion claymation scene — real plasticine clay characters and set with visible thumbprints, tool marks, seam lines and a soft matte sheen, miniature studio photography with shallow depth of field and warm practical lighting, handmade Aardman-quality charm. NOT digital CG, NOT a 3D render, NOT 2D illustration — a physically sculpted miniature scene photographed on a real stop-motion stage
```

  - **[GPT Image 2](https://nodaro.ai/docs/models/image/gpt-image-2)** (Pick) · 89 s · 1672×941. Storm, lighthouse, clay all there ![GPT Image 2's result for “Claymation Scene”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r2/claymation-scene-gpt-image-2.webp) Full-size original: https://cdn.nodaro.ai/images/9479f7b9-b31b-4574-9ed3-3afd0cb666c6.png
  - **[Nano Banana 2.1](https://nodaro.ai/docs/models/image/nano-banana-2-1)** · 55 s · 1376×768 ![Nano Banana 2.1's result for “Claymation Scene”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r2/claymation-scene-nano-banana-2-1.webp) Full-size original: https://cdn.nodaro.ai/images/298b8a78-22e4-4ae4-816f-d58d33894645.png
  - **[Nano Banana Pro](https://nodaro.ai/docs/models/image/nano-banana-pro)** (Weak) · 36 s · 1376×768. Flat light, no storm ![Nano Banana Pro's result for “Claymation Scene”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r2/claymation-scene-nano-banana-pro.webp) Full-size original: https://cdn.nodaro.ai/images/71207246-b8d4-4874-a3a4-fbe1035bbde4.png

### Studio Portrait
```
studio portrait of a confident woman with sleek dark hair, 85mm lens at f/2.8, shallow depth of field, soft octabox key with gentle fill and a subtle hair light, clean seamless gray backdrop, tight head-and-shoulders framing, polished editorial mood with a neutral true-to-life palette, natural skin texture, catchlights and tack-sharp eyes
```

  - **[GPT Image 2](https://nodaro.ai/docs/models/image/gpt-image-2)** · 79 s · 1086×1448. Drawn at 3:4 — it has no 4:5 ![GPT Image 2's result for “Studio Portrait”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r2/studio-portrait-gpt-image-2.webp) Full-size original: https://cdn.nodaro.ai/images/56bc633e-10be-4619-9805-e6d55b44ded0.png
  - **[Nano Banana 2.1](https://nodaro.ai/docs/models/image/nano-banana-2-1)** (Pick) · 27 s · 928×1152. Natural studio light at the preset's 4:5 ![Nano Banana 2.1's result for “Studio Portrait”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r2/studio-portrait-nano-banana-2-1.webp) Full-size original: https://cdn.nodaro.ai/images/57bc0512-0bd3-4332-8b35-da7da6f302e8.png
  - **[Nano Banana Pro](https://nodaro.ai/docs/models/image/nano-banana-pro)** · 35 s · 928×1152 ![Nano Banana Pro's result for “Studio Portrait”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r2/studio-portrait-nano-banana-pro.webp) Full-size original: https://cdn.nodaro.ai/images/8c618e33-78e0-4fc0-9e89-a3ad0266c781.png

### Three-Age Triptych
```
Single photorealistic studio composite of the SAME person from the reference image as a three-age triptych, shoulder-to-shoulder left to right: around age 10, around age 40, and around age 75. Editorial portrait photography, 85mm lens at f/2.8, each figure framed chest-up, facing camera with the same calm expression. CRITICAL identity lock — same eyes, eye color, brow, nose, ears, jawline and facial bone structure across all three, unmistakably the same person; age ONLY via skin texture, fine lines, hairline, hair color (darker in youth, greying with age), and a slight shift in facial fullness. Age-appropriate clothing per stage. Even soft cinematic studio key light with gentle rim separation, clean dark-grey gradient seamless backdrop, refined neutral grade, natural skin tones, tack-sharp facial detail; the three figures fill the frame.
```

  
![The starting photo: a young woman looking straight at the camera](https://nodaro.ai/docs-media/research/image-model-bakeoff/source/src-personA.webp)

The starting photo

  - **[GPT Image 2](https://nodaro.ai/docs/models/image/gpt-image-2)** (Pick) · 107 s · 1086×1448. One person, three ages ![GPT Image 2's result for “Three-Age Triptych”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r2/age-progression-3-gpt-image-2.webp) Full-size original: https://cdn.nodaro.ai/images/93111160-dcc9-48c6-8db2-f22b909d58f2.png
  - **[Nano Banana 2.1](https://nodaro.ai/docs/models/image/nano-banana-2-1)** · 201 s · 896×1200 ![Nano Banana 2.1's result for “Three-Age Triptych”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r2/age-progression-3-nano-banana-2-1.webp) Full-size original: https://cdn.nodaro.ai/images/b5261e9b-6d3c-4c06-ae70-9a92e5a96325.png
  - **[Nano Banana Pro](https://nodaro.ai/docs/models/image/nano-banana-pro)** · 40 s · 896×1200 ![Nano Banana Pro's result for “Three-Age Triptych”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r2/age-progression-3-nano-banana-pro.webp) Full-size original: https://cdn.nodaro.ai/images/61bfc940-d7a2-4060-af58-ed3f5a1417bd.png

## Round 3: Seedream 5 Pro and Seedream 5 Flash
We ran every prompt of rounds 1 and 2 again on Seedream 5 Pro and Seedream 5 Flash, and compared each result with that prompt's earlier pick: a tie, close to it, worse, or a fail.

| Model | Tie | Close | Worse | Fail | Median time | Credits per image |
| --- | --- | --- | --- | --- | --- | --- |
| [Seedream 5 Flash](https://nodaro.ai/docs/models/image/seedream-5-flash) | 7 | 6 | 4 | 1 | 45 s | 11 |
| [Seedream 5 Pro](https://nodaro.ai/docs/models/image/seedream-5-pro) | 5 | 3 | 9 | 1 | 100 s | 20–66 |

Neither model beat an earlier pick on any prompt. Seedream 5 Flash came closest, at a fraction of the price: a good draft model, not a quality default.

#### Swap the clothes
The same prompt as [Swap the clothes](#edit-clothes).

  - **[GPT Image 2.5 Flare](https://nodaro.ai/docs/models/image/gpt-image-2-5-flare)** (Earlier pick) · 50 s · 2368×1776 ![GPT Image 2.5 Flare's result for “Swap the clothes”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/edit-clothes-D.webp) Full-size original: https://cdn.nodaro.ai/images/90fe31a1-a581-47d0-84ff-97c3fa7059d5.png
  - **[Seedream 5 Pro](https://nodaro.ai/docs/models/image/seedream-5-pro-i2i)** (Tie) · 90 s · 2304×1728. Identity, hair and pose kept; the jacket swap is as clean as Flare's. ![Seedream 5 Pro's result for “Swap the clothes”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r3/edit-clothes-pro.webp) Full-size original: https://cdn.nodaro.ai/images/f3077ba0-d740-4c3b-95bd-d6894571a305.png
  - **[Seedream 5 Flash](https://nodaro.ai/docs/models/image/seedream-5-flash-i2i)** (Tie) · 52 s · 2368×1776. Identity, hair and pose kept; jacket swap as clean as Flare's. ![Seedream 5 Flash's result for “Swap the clothes”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r3/edit-clothes-flash.webp) Full-size original: https://cdn.nodaro.ai/images/b67bbe6f-c827-471d-9ab8-ceca3e00be93.png

#### Replace the background
The same prompt as [Replace the background](#edit-background).

  - **[GPT Image 2.5 Sunburst](https://nodaro.ai/docs/models/image/gpt-image-2-5-sunburst)** (Earlier pick) · 56 s · 1776×2368 ![GPT Image 2.5 Sunburst's result for “Replace the background”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/edit-background-A.webp) Full-size original: https://cdn.nodaro.ai/images/a0b42c94-a29c-4f79-9480-9294b7a6c03c.png
  - **[Seedream 5 Pro](https://nodaro.ai/docs/models/image/seedream-5-pro-i2i)** (Tie) · 130 s · 1728×2304. Same man, glasses and polo kept; office background as convincing as Sunburst's. ![Seedream 5 Pro's result for “Replace the background”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r3/edit-background-pro.webp) Full-size original: https://cdn.nodaro.ai/images/6f41d49b-f574-4630-b82d-fe46f190f6f2.png
  - **[Seedream 5 Flash](https://nodaro.ai/docs/models/image/seedream-5-flash-i2i)** (Tie) · 52 s · 1776×2368. Same man, glasses and polo; a convincing office, on a par with Sunburst. ![Seedream 5 Flash's result for “Replace the background”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r3/edit-background-flash.webp) Full-size original: https://cdn.nodaro.ai/images/7f3297c6-169b-4471-b378-26d51fe96165.png

#### Restage a product photo
The same prompt as [Restage a product photo](#edit-product).

  - **[GPT Image 2.5 Flare](https://nodaro.ai/docs/models/image/gpt-image-2-5-flare)** (Earlier pick) · 86 s · 2368×1776 ![GPT Image 2.5 Flare's result for “Restage a product photo”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/edit-product-A.webp) Full-size original: https://cdn.nodaro.ai/images/d6720ab3-92ec-4a69-bf5f-cdc27f3c6bdf.png
  - **[Seedream 5 Pro](https://nodaro.ai/docs/models/image/seedream-5-pro-i2i)** (Worse) · 162 s · 2304×1728. Headphones kept, but it re-framed the scene and dropped the plant and the book; Flare kept the setting. ![Seedream 5 Pro's result for “Restage a product photo”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r3/edit-product-pro.webp) Full-size original: https://cdn.nodaro.ai/images/f411ec9d-8300-4d78-a0e3-4fe627ee1e2e.png
  - **[Seedream 5 Flash](https://nodaro.ai/docs/models/image/seedream-5-flash-i2i)** (Close) · 57 s · 2368×1776. Headphones kept exactly; the setting is simplified (no plant or book), but it reads better than Pro's. ![Seedream 5 Flash's result for “Restage a product photo”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r3/edit-product-flash.webp) Full-size original: https://cdn.nodaro.ai/images/e4630077-d566-4368-bedb-e18df174bff6.png

#### Put the person in a new scene
The same prompt as [Put the person in a new scene](#id-scene).

  - **[GPT Image 2.5 Flare](https://nodaro.ai/docs/models/image/gpt-image-2-5-flare)** (Earlier pick) · 51 s · 1776×2368 ![GPT Image 2.5 Flare's result for “Put the person in a new scene”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/id-scene-C.webp) Full-size original: https://cdn.nodaro.ai/images/fa2b85c2-c609-474a-bbde-033d951bbcd0.png
  - **[Seedream 5 Pro](https://nodaro.ai/docs/models/image/seedream-5-pro)** (Close) · 98 s · 1776×2352. Identity holds and the café reads; Flare's frame is warmer and has the Paris cue (Eiffel Tower). ![Seedream 5 Pro's result for “Put the person in a new scene”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r3/id-scene-pro.webp) Full-size original: https://cdn.nodaro.ai/images/3b546c40-640b-4231-b8e6-da6723585e33.png
  - **[Seedream 5 Flash](https://nodaro.ai/docs/models/image/seedream-5-flash)** (Close) · 46 s · 1776×2368. Identity holds and the Paris café reads; Flare's frame is still warmer. ![Seedream 5 Flash's result for “Put the person in a new scene”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r3/id-scene-flash.webp) Full-size original: https://cdn.nodaro.ai/images/e5506410-5de0-4ed9-8c66-f0c838fee4bb.png

#### Character reference grid
The same prompt as [Character reference grid](#id-grid).

  - **[GPT Image 2](https://nodaro.ai/docs/models/image/gpt-image-2)** (Earlier pick) · 79 s · 2448×3264 ![GPT Image 2's result for “Character reference grid”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/id-grid-A.webp) Full-size original: https://cdn.nodaro.ai/images/7d73dded-44c2-472c-a4c2-ef3366e78a8e.png
  - **[Seedream 5 Pro](https://nodaro.ai/docs/models/image/seedream-5-pro)** (Worse) · 162 s · 1728×2304. Face holds, but the outfit drifts (cold-shoulder top, burgundy shoes) across the four views. ![Seedream 5 Pro's result for “Character reference grid”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r3/id-grid-pro.webp) Full-size original: https://cdn.nodaro.ai/images/0f286bb3-843d-42a3-b83a-45cca5da9d75.png
  - **[Seedream 5 Flash](https://nodaro.ai/docs/models/image/seedream-5-flash)** (Worse) · 52 s · 1776×2368. Face holds, but the outfit changed to a dress — the identity grid has to keep the clothes. ![Seedream 5 Flash's result for “Character reference grid”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r3/id-grid-flash.webp) Full-size original: https://cdn.nodaro.ai/images/3832d4e6-caef-417c-9792-f2d663c906ce.png

#### Two-person cast sheet
The same prompt as [Two-person cast sheet](#id-cast).

  - **[Nano Banana 2.1](https://nodaro.ai/docs/models/image/nano-banana-2-1)** (Earlier pick) · 98 s · 3584×4800 ![Nano Banana 2.1's result for “Two-person cast sheet”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/id-cast-A.webp) Full-size original: https://cdn.nodaro.ai/images/969d9f79-7684-4a89-958f-364d5494b12c.png
  - **[Seedream 5 Pro](https://nodaro.ai/docs/models/image/seedream-5-pro)** (Worse) · 173 s · 1776×2352. Both people recognisable, but Maya's top changed and only one name label rendered. ![Seedream 5 Pro's result for “Two-person cast sheet”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r3/id-cast-pro.webp) Full-size original: https://cdn.nodaro.ai/images/6ac4e0b7-e400-436d-ad81-5efc2f3fbf6a.png
  - **[Seedream 5 Flash](https://nodaro.ai/docs/models/image/seedream-5-flash)** (Fail) · 44 s · 1776×2368. Mixed the two people up: Maya's row ends on David's body and David's row is Maya. ![Seedream 5 Flash's result for “Two-person cast sheet”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r3/id-cast-flash.webp) Full-size original: https://cdn.nodaro.ai/images/9335e80a-ed72-4a66-8fd7-5775d4acbeff.png

#### Rainy street at night
The same prompt as [Rainy street at night](#t2i-street).

  - **[Nano Banana 2.1](https://nodaro.ai/docs/models/image/nano-banana-2-1)** (Earlier pick) · 97 s · 2752×1536 ![Nano Banana 2.1's result for “Rainy street at night”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/t2i-street-B.webp) Full-size original: https://cdn.nodaro.ai/images/811306c9-9a46-426b-9f29-c70e956a9276.png
  - **[Seedream 5 Pro](https://nodaro.ai/docs/models/image/seedream-5-pro)** (Worse) · 89 s · 2720×1530. Good night mood, but a plainer frame with less signage and detail than Nano Banana 2.1. ![Seedream 5 Pro's result for “Rainy street at night”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r3/t2i-street-pro.webp) Full-size original: https://cdn.nodaro.ai/images/3a2e8722-75ea-4b62-a1ba-e97635dbb0fc.png
  - **[Seedream 5 Flash](https://nodaro.ai/docs/models/image/seedream-5-flash)** (Worse) · 71 s · 2816×1584. Moody, but the ramen shop barely shows and the frame is mostly dark street. ![Seedream 5 Flash's result for “Rainy street at night”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r3/t2i-street-flash.webp) Full-size original: https://cdn.nodaro.ai/images/4a704a8a-d7e2-4545-b00d-f4ea27014816.png

#### Product photo of a mug
The same prompt as [Product photo of a mug](#t2i-product).

  - **[GPT Image 2](https://nodaro.ai/docs/models/image/gpt-image-2)** (Earlier pick) · 68 s · 2048×2048 ![GPT Image 2's result for “Product photo of a mug”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/t2i-product-A.webp) Full-size original: https://cdn.nodaro.ai/images/eec22388-4123-489d-92e2-4b69cb005dce.png
  - **[Seedream 5 Pro](https://nodaro.ai/docs/models/image/seedream-5-pro)** (Worse) · 95 s · 2048×2048. Clean, but the mug sits small and the shot is flatter than GPT Image 2's. ![Seedream 5 Pro's result for “Product photo of a mug”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r3/t2i-product-pro.webp) Full-size original: https://cdn.nodaro.ai/images/85434968-343e-4b94-9b98-f74decf2a0b5.png
  - **[Seedream 5 Flash](https://nodaro.ai/docs/models/image/seedream-5-flash)** (Close) · 46 s · 2048×2048. A good, well-lit mug; GPT Image 2's is a touch crisper. ![Seedream 5 Flash's result for “Product photo of a mug”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r3/t2i-product-flash.webp) Full-size original: https://cdn.nodaro.ai/images/e5206843-bc34-4e29-ab80-6a23b0d94f04.png

#### Watercolor picture-book illustration
The same prompt as [Watercolor picture-book illustration](#t2i-illustration).

  - **[Nano Banana 2.1](https://nodaro.ai/docs/models/image/nano-banana-2-1)** (Earlier pick) · 59 s · 2400×1792 ![Nano Banana 2.1's result for “Watercolor picture-book illustration”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/t2i-illustration-D.webp) Full-size original: https://cdn.nodaro.ai/images/c0a5aeaf-7af0-46ea-8f3c-e77b3a079d57.png
  - **[Seedream 5 Pro](https://nodaro.ai/docs/models/image/seedream-5-pro)** (Tie) · 101 s · 2304×1728. Charming watercolour, on a par with Nano Banana 2.1. ![Seedream 5 Pro's result for “Watercolor picture-book illustration”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r3/t2i-illustration-pro.webp) Full-size original: https://cdn.nodaro.ai/images/aa1be8de-3f3e-4ed5-9517-5df500ca85b5.png
  - **[Seedream 5 Flash](https://nodaro.ai/docs/models/image/seedream-5-flash)** (Tie) · 46 s · 2368×1776. Lovely watercolour, on a par with Nano Banana 2.1. ![Seedream 5 Flash's result for “Watercolor picture-book illustration”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r3/t2i-illustration-flash.webp) Full-size original: https://cdn.nodaro.ai/images/8bc22225-be0e-457e-b583-e3009554d07c.png

#### Concert poster with exact text
The same prompt as [Concert poster with exact text](#type-poster).

  - **[GPT Image 2.5 Flare](https://nodaro.ai/docs/models/image/gpt-image-2-5-flare)** (Earlier pick) · 59 s · 1776×2368 ![GPT Image 2.5 Flare's result for “Concert poster with exact text”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/type-poster-B.webp) Full-size original: https://cdn.nodaro.ai/images/2d6dfa69-87c3-47f0-8322-a8f17f35753a.png
  - **[Seedream 5 Pro](https://nodaro.ai/docs/models/image/seedream-5-pro)** (Worse) · 120 s · 1728×2304. Every word spelled right, but the layout is weaker: no hero image, washed-out title. ![Seedream 5 Pro's result for “Concert poster with exact text”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r3/type-poster-pro.webp) Full-size original: https://cdn.nodaro.ai/images/3c5666a6-12c5-4768-baa6-05f1ff5ec1db.png
  - **[Seedream 5 Flash](https://nodaro.ai/docs/models/image/seedream-5-flash)** (Worse) · 44 s · 1776×2368. Words right, but a thin layout with a line break in the date line; Flare's poster is far stronger. ![Seedream 5 Flash's result for “Concert poster with exact text”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r3/type-poster-flash.webp) Full-size original: https://cdn.nodaro.ai/images/402b4742-c267-4078-b468-6b1ea19d5e3d.png

#### Infographic with exact labels
The same prompt as [Infographic with exact labels](#type-infographic).

  - **[GPT Image 2](https://nodaro.ai/docs/models/image/gpt-image-2)** (Earlier pick) · 80 s · 2016×2688 ![GPT Image 2's result for “Infographic with exact labels”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r1/type-infographic-D.webp) Full-size original: https://cdn.nodaro.ai/images/a5cd4e9e-11de-4e8a-982e-ed531c2c2e37.png
  - **[Seedream 5 Pro](https://nodaro.ai/docs/models/image/seedream-5-pro)** (Close) · 125 s · 1776×2352. Clean steps with the right labels, but each label is printed twice. ![Seedream 5 Pro's result for “Infographic with exact labels”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r3/type-infographic-pro.webp) Full-size original: https://cdn.nodaro.ai/images/92f569f2-ead0-47f0-b842-d595a2be4cbf.png
  - **[Seedream 5 Flash](https://nodaro.ai/docs/models/image/seedream-5-flash)** (Tie) · 40 s · 1776×2368. Clean, correct steps and labels with arrows — as good as GPT Image 2 (and better than Pro). ![Seedream 5 Flash's result for “Infographic with exact labels”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r3/type-infographic-flash.webp) Full-size original: https://cdn.nodaro.ai/images/910c358b-0d93-442b-9717-57870a190e76.png

#### Character Board
The same prompt as [Character Board](#character-board).

  - **[GPT Image 2](https://nodaro.ai/docs/models/image/gpt-image-2)** (Earlier pick) · 111 s · 2048×1152 ![GPT Image 2's result for “Character Board”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r2/character-board-gpt-image-2.webp) Full-size original: https://cdn.nodaro.ai/images/fd06a91f-1baf-401a-8fa8-d3fdc3c22b06.png
  - **[Seedream 5 Pro](https://nodaro.ai/docs/models/image/seedream-5-pro)** (Worse) · 191 s · 1312×736. Strong look, but the small text is gibberish and the outfit varies between panels. ![Seedream 5 Pro's result for “Character Board”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r3/character-board-pro.webp) Full-size original: https://cdn.nodaro.ai/images/5ce01c64-b01c-4868-a4b5-12fff4fc3ad0.png
  - **[Seedream 5 Flash](https://nodaro.ai/docs/models/image/seedream-5-flash)** (Worse) · 45 s · 1424×800. Panels exist, but the labels are garbled and several panels are wrong. ![Seedream 5 Flash's result for “Character Board”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r3/character-board-flash.webp) Full-size original: https://cdn.nodaro.ai/images/479d9663-3903-4edf-83b5-0a91c215a44e.png

#### Character Sheet · Expressions ×6
The same prompt as [Character Sheet · Expressions ×6](#character-expressions-6).

  - **[Nano Banana 2.1](https://nodaro.ai/docs/models/image/nano-banana-2-1)** (Earlier pick) · 49 s · 1200×896 ![Nano Banana 2.1's result for “Character Sheet · Expressions ×6”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r2/character-expressions-6-nano-banana-2-1.webp) Full-size original: https://cdn.nodaro.ai/images/4b27dd4f-a785-4e11-9c0b-78524ca1bd55.png
  - **[Seedream 5 Pro](https://nodaro.ai/docs/models/image/seedream-5-pro)** (Close) · 97 s · 1184×896. Six clear expressions with correct labels, but drawn semi-illustrated, not photographic. ![Seedream 5 Pro's result for “Character Sheet · Expressions ×6”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r3/character-expressions-6-pro.webp) Full-size original: https://cdn.nodaro.ai/images/f291cd29-3080-4da3-90d9-0cc7bba3a4ab.png
  - **[Seedream 5 Flash](https://nodaro.ai/docs/models/image/seedream-5-flash)** (Close) · 27 s · 1152×864. Six clear, labelled expressions, consistent face; slightly illustrated rather than photographic. ![Seedream 5 Flash's result for “Character Sheet · Expressions ×6”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r3/character-expressions-6-flash.webp) Full-size original: https://cdn.nodaro.ai/images/8d7e7c9b-e7be-4ef2-972c-966904ff70fd.png

#### Product Shot (white bg)
The same prompt as [Product Shot (white bg)](#product-shot).

  - **[GPT Image 2](https://nodaro.ai/docs/models/image/gpt-image-2)** (Earlier pick) · 65 s · 1024×1024 ![GPT Image 2's result for “Product Shot (white bg)”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r2/product-shot-gpt-image-2.webp) Full-size original: https://cdn.nodaro.ai/images/afe4a2c5-7972-4434-b873-c5dbdc81e87c.png
  - **[Seedream 5 Pro](https://nodaro.ai/docs/models/image/seedream-5-pro)** (Worse) · 79 s · 1024×1024. Product small in frame; GPT Image 2 fills the frame with a crisper hero shot. ![Seedream 5 Pro's result for “Product Shot (white bg)”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r3/product-shot-pro.webp) Full-size original: https://cdn.nodaro.ai/images/4b4ff51e-8703-4a55-bc96-c797a9039eb1.png
  - **[Seedream 5 Flash](https://nodaro.ai/docs/models/image/seedream-5-flash)** (Close) · 33 s · 1024×1024. Clean product, smaller in frame than GPT Image 2's hero shot. ![Seedream 5 Flash's result for “Product Shot (white bg)”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r3/product-shot-flash.webp) Full-size original: https://cdn.nodaro.ai/images/0475a95a-f2f8-485c-8596-935f8629bade.png

#### Movie Poster
The same prompt as [Movie Poster](#movie-poster).

  - **[GPT Image 2.5 Sunburst](https://nodaro.ai/docs/models/image/gpt-image-2-5-sunburst)** (Earlier pick) · 94 s · 1024×1536 ![GPT Image 2.5 Sunburst's result for “Movie Poster”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r2/movie-poster-gpt-image-2-5-sunburst.webp) Full-size original: https://cdn.nodaro.ai/images/08d21907-b434-4fa5-af60-a4a029aabc52.png
  - **[Seedream 5 Pro](https://nodaro.ai/docs/models/image/seedream-5-pro)** (Tie) · 128 s · 832×1248. Strong poster and readable title; on a par with Sunburst. ![Seedream 5 Pro's result for “Movie Poster”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r3/movie-poster-pro.webp) Full-size original: https://cdn.nodaro.ai/images/94d06fa6-9444-4274-a87a-06c7bf8ebe49.png
  - **[Seedream 5 Flash](https://nodaro.ai/docs/models/image/seedream-5-flash)** (Tie) · 38 s · 832×1248. A strong poster with a bold, legible title; on a par with Sunburst. ![Seedream 5 Flash's result for “Movie Poster”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r3/movie-poster-flash.webp) Full-size original: https://cdn.nodaro.ai/images/68205e8e-9a33-4e3d-84dc-b616925a0372.png

#### Claymation Scene
The same prompt as [Claymation Scene](#claymation-scene).

  - **[GPT Image 2](https://nodaro.ai/docs/models/image/gpt-image-2)** (Earlier pick) · 89 s · 1672×941 ![GPT Image 2's result for “Claymation Scene”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r2/claymation-scene-gpt-image-2.webp) Full-size original: https://cdn.nodaro.ai/images/9479f7b9-b31b-4574-9ed3-3afd0cb666c6.png
  - **[Seedream 5 Pro](https://nodaro.ai/docs/models/image/seedream-5-pro)** (Worse) · 85 s · 1312×736. Good clay texture, but a sparser scene (no lighthouse or storm) than GPT Image 2's. ![Seedream 5 Pro's result for “Claymation Scene”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r3/claymation-scene-pro.webp) Full-size original: https://cdn.nodaro.ai/images/65ecc4ea-1d16-4c53-a3d4-3b29db3954e4.png
  - **[Seedream 5 Flash](https://nodaro.ai/docs/models/image/seedream-5-flash)** (Close) · 33 s · 1424×800. Good clay, lighthouse and storm are there; less dramatic light than GPT Image 2. ![Seedream 5 Flash's result for “Claymation Scene”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r3/claymation-scene-flash.webp) Full-size original: https://cdn.nodaro.ai/images/fb4348fd-a1b1-4dc7-828c-d5d0c14a0d42.png

#### Studio Portrait
The same prompt as [Studio Portrait](#studio-portrait).

  - **[Nano Banana 2.1](https://nodaro.ai/docs/models/image/nano-banana-2-1)** (Earlier pick) · 27 s · 928×1152 ![Nano Banana 2.1's result for “Studio Portrait”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r2/studio-portrait-nano-banana-2-1.webp) Full-size original: https://cdn.nodaro.ai/images/57bc0512-0bd3-4332-8b35-da7da6f302e8.png
  - **[Seedream 5 Pro](https://nodaro.ai/docs/models/image/seedream-5-pro)** (Tie) · 69 s · 896×1184. Polished studio light, on a par with Nano Banana 2.1. ![Seedream 5 Pro's result for “Studio Portrait”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r3/studio-portrait-pro.webp) Full-size original: https://cdn.nodaro.ai/images/49cacbe3-743d-4a19-985a-5a980f00d19b.png
  - **[Seedream 5 Flash](https://nodaro.ai/docs/models/image/seedream-5-flash)** (Tie) · 32 s · 864×1152. Natural, polished studio portrait, on a par with Nano Banana 2.1. ![Seedream 5 Flash's result for “Studio Portrait”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r3/studio-portrait-flash.webp) Full-size original: https://cdn.nodaro.ai/images/50438fae-f216-4a61-a13a-4402c0a74656.png

#### Three-Age Triptych
The same prompt as [Three-Age Triptych](#age-progression-3).

  - **[GPT Image 2](https://nodaro.ai/docs/models/image/gpt-image-2)** (Earlier pick) · 107 s · 1086×1448 ![GPT Image 2's result for “Three-Age Triptych”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r2/age-progression-3-gpt-image-2.webp) Full-size original: https://cdn.nodaro.ai/images/93111160-dcc9-48c6-8db2-f22b909d58f2.png
  - **[Seedream 5 Pro](https://nodaro.ai/docs/models/image/seedream-5-pro)** (Fail) · 77 s · 864×1152. The oldest face is clearly a different person (different ethnicity), so the progression breaks. ![Seedream 5 Pro's result for “Three-Age Triptych”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r3/age-progression-3-pro.webp) Full-size original: https://cdn.nodaro.ai/images/9af5596d-4da2-490c-b3fc-f2909fea1f44.png
  - **[Seedream 5 Flash](https://nodaro.ai/docs/models/image/seedream-5-flash)** (Tie) · 32 s · 864×1152. One recognisable person across the three ages — as good as GPT Image 2, where Pro failed. ![Seedream 5 Flash's result for “Three-Age Triptych”](https://nodaro.ai/docs-media/research/image-model-bakeoff/r3/age-progression-3-flash.webp) Full-size original: https://cdn.nodaro.ai/images/259d1f49-67dc-4ac9-9e01-488bb809bb11.png

## What quality costs
- **Speed:** GPT Image 2 had the slowest median in round 1, 97 seconds, against 51 to 68 seconds for the other four.
- **Ratios:** GPT Image 2 draws only auto, 1:1, 16:9, 9:16, 4:3 and 3:4. For 2:3, GPT Image 2.5 Sunburst made the winning movie poster, and for 4:5, Nano Banana 2.1 made the winning studio portrait.
- **Clothing:** the GPT models refuse some clothing, such as swimwear.
- **Price:** at 2K, GPT Image 2 costs 42 credits an image, against 22 for Nano Banana 2.1, 28 for GPT Image 2.5 Flare and Sunburst, and 11 for Seedream 5 Flash.

## Try them yourself
Every model in this test runs in Nodaro: on the [Generate Image](https://nodaro.ai/docs/nodes/image/generate-image) node for a new image, and on the [Modify Image](https://nodaro.ai/docs/nodes/image/modify-image) node for an edit. To run your own prompt on several models at once, connect one **Text** node to several image nodes and choose a different model in each. [Choosing a model](https://nodaro.ai/docs/guides/choosing-models) compares every image model in Nodaro.

## Frequently asked questions

### Which AI image model is best for editing photos?

In our test, GPT Image 2.5 Flare. It won two of the three edits and kept the pose, the framing and the product design on all three. GPT Image 2.5 Sunburst won the third edit, a background swap.

### Which AI image model keeps a character consistent?

GPT Image 2 made the best reference grid, kept both people right on the cast sheet, and won the character board and the age progression. Nano Banana 2.1 won the labelled cast sheet and the expression sheet.

### Which AI image model writes text best?

GPT Image 2 drew the cleanest infographic and GPT Image 2.5 Flare the best concert poster. Nano Banana Pro rendered the poster as a skewed photo of a print and returned a nearly empty infographic.

### Is Nano Banana Pro still worth using?

Not for these jobs. It won none of the 18 prompts and got 8 weak marks in 11 prompts, including a broken grid layout, a dropped name label and changed outfits.

### Is Seedream 5 Flash good for its price?

Yes, as a draft model. At 11 credits an image, it tied the earlier pick on 7 of 18 prompts and came close on 6, in about 45 seconds. It beat no earlier pick and mixed up the two people on the cast sheet.

### How were the winners picked?

Claude, an AI model, looked at each prompt's results without the model names and chose the pick and any weak results. Each model made one image per prompt, so read a single result as a hint and the pattern across prompts as the finding.