Lazza Studio | Best Practices

It is recommended that you generate product materials following the methods suggested in this best practices guide.

1. Model Strengths

Mode

Corresponding Model

Pricing

Advantages & Disadvantages

Turbo Mode


 

Nanobanana


 

6 Craft Points per Image


 

Advantages:

1. Fast and high-quality

2. Suitable for generating white-background images, recolored images, etc.


 

Disadvantages:

1. Only supports generating 1K images

2. May not perform well for creative or imaginative poster designs

3. Stricter restrictions on sensitive content (e.g., swimwear, clothing with certain levels of exposure)

4. Currently does not support generating multiple different images from a single prompt (e.g., red, blue, yellow)

Standard Commercial Mode


 

Seedream 4.5


 

6 Craft Points per Image

Advantages:

1. Fast and high-quality

2. No restrictions on sensitive content (recommended for generating swimwear and related clothing materials)

3. Performs well for creative content and poster designs

4. Supports generating multiple different images from a single prompt (e.g., red, blue, yellow)

Disadvantages:

May not maintain full consistency with the original image (e.g., textures may change, faces may vary, poses may differ)

Professional Studio


 

Nanobanana Pro

1K / 2K: 20 Craft Points per Image

4K: 30 Craft Points per Image


 

Advantages:

Produces the highest quality images, whether for basic recoloring or complex combinations and creative poster designs

Disadvantages:

1. Slower generation speed

2. Nanobanana Pro’s global resource pool is highly utilized, which may result in temporary inability to generate images

3. If you experience long waits while generating images in "Professional Studio," you can click "AI Product Image" to start a new session and try other generation tasks

4. Stricter restrictions on sensitive content (e.g., swimwear, clothing with certain levels of exposure)

5. Currently does not support generating multiple different images from a single prompt (e.g., red, blue, yellow)


2. Best Practice Path

Module

Product Interface

Best Practice Path

Input Before Image Generation


 


 


 

Input Box Configuration

1. Reference Images

  • It is recommended to upload reference images each time you generate images, such as your product or images requiring secondary creation.

2. Configuration Options

 Resolution / Aspect Ratio:

  • If not selected, images will be generated according to the default resolution & aspect ratio for each mode.
  • Professional Studio officially recommends using 2K.
  • If both the input box and the configuration options specify different resolutions or aspect ratios, the configuration options take precedence.
  • Default Resolution & Aspect Ratio by Mode:
    • Turbo Mode: 1:1, 1K
    • Standard Commercial Mode: 1:1, 2K
    • Professional Studio: 1:1, 2K

3. Number of Images

  • It is recommended to generate multiple images at once. Since generative models have inherent variability, generating multiple images reduces the chance of random results and increases the likelihood of producing images that match your expectations.

4. Optimize for Me

  • Strongly recommended feature.
  • After entering your generation idea, click "Optimize for Me." Lazza Studio will use an AI language model to refine your prompt, producing higher-quality generation instructions and improving the final image quality.

5. Generation Instructions

  • When uploading multiple reference images, clearly indicate which parts of Image 1, Image 2, etc., you want to keep consistent or replace.
  • If you require consistency in facial features, textures, or other elements, describe these requirements clearly in the generation instructions.


 

After generating images, secondary editing of the generated images is required


 


 

Secondary Editing of Generated Images

It is recommended to hover your mouse over the image you want to edit and use the "Reference Image" or "Copy Image" function to re-add the image to the input box as a reference for secondary editing.

  • This effectively avoids interference from the conversation context, preventing confusion during image generation.
生成模特 

Generate Models

It is recommended to upload only a front-facing photo of the model. This ensures facial features remain consistent in subsequent poster or outfit generation.

Was this article helpful?
0 out of 0 found this helpful

Comments

0 comments

Please sign in to leave a comment.

Live chat
Reset