Text to Image allows you to generate an AI image using a text prompt, which you can then use as the input for Meshy's Image to 3D feature. This feature unlocks the full power of Meshy and Nano Banana Pro, allowing you to modify images iteratively with a high level of artistic control until you are satisfied and ready to turn your final image into a 3D model. The effect is similar to modifying the 3D mesh directly using text prompt - you just do it in the image step first before generating our 3D model.
You can also use our AI Helper to create an AI image generated from a text prompt and modify the image iteratively within the Helper until you are satisfied, before generating your model from the final AI image.
Note that you will need to use the helper inside the "Model" tab (On the left) for this, and ensure "Create Image" at the bottom is selected.
🖼️ Using Nano Banana Pro in Image Workspace
The Prompt field lets you generate and refine AI images using Nano Banana Pro.
You can use it in two main ways:
✨ 1. Generate an AI Image from Text
Simply describe what you want to create.
Example:
“Stylized Low-Poly Dragon Sword, with Ruby Gems”
Nano Banana Pro will generate an AI image based on your text prompt.
You can generate multiple variations and iterate until satisfied.
You can write in your native language.
You can also use the AI prompt helper in the AI Image tab to generate text prompts to use.
🖼 Choose Aspect Ratio
Select the shape of your generated image:
1:1 – Square (great for 3D conversion)
16:9 – Landscape
9:16 – Portrait
4:3
3:4
For Image to 3D workflows, 1:1 is often recommended.
🔢 Select Number of Images
Choose how many variations you want to generate (1-8)
Generating multiple images helps you:
Compare variations
Pick the strongest composition
Refine your direction
Note: More images = more credits used.
🔁 Multi-View (Optional)
Toggle Multi-View if you want multiple angles of the same subject.
This is useful for:
Character design
Object modeling
Improving 3D reconstruction accuracy
🧍 Select Pose (If Applicable)
Choose a pose option:
None – Default generation
A-Pose
T-Pose
For characters intended for 3D modeling or rigging:
A-Pose or T-Pose is recommended.
For environment or prop concepts:
Keep it set to None.
🖼 Add Image References (Optional)
You can use up to 5 reference images. Either directly uploaded or sent from previously generated AI images in the workspace.
Ways to upload:
Click
Drag & Drop
Paste
Supported formats:
PNG
JPG / JPEG
WEBP
Max size: 20MB per image
⏱ Generation Details
At the bottom, you’ll see:
Estimated generation time (e.g., 15 seconds)
Credit cost (e.g., 9 × 4 = 36)
This updates based on:
Number of images
Selected settings
🚀 Click Generate
Press the Generate button to create your image(s).
Once generated, you can:
Select your preferred image
Refine it further
Use it as input for Image to 3D
Use it as an input for future reference for AI Images.
Iterate until satisfied, then select "Image to 3D" to create your 3D model with it!
🔁 2. Modify a Reference Image
Upload a reference image and tell Nano Banana Pro how to change it:
“Make this low-poly.”
“Simplify for 3D printing.”
“Change the material to gold with cracks in it”
“Remove background and change to front-facing angle”
The AI will reinterpret your reference based on your instructions.
TIP: You can also send a previously generated AI image to be used as reference by selecting the three dot menu below it.
🧩 3. Combine Multiple Reference Images
Upload multiple images and guide how they should merge:
“Use the shape from Image 1 and the style from Image 2.”
“Combine this armor design with that pose.”
“Blend these into one cohesive prop.”
This allows for controlled design iteration and concept merging.
If you find any interesting workflows or tips for our AI Image feature, let us know in the Meshy Community Discord!