Back to Nodes
Nano Banana 2

Nano Banana 2

Official

Google's fast image generation and editing model with fal.ai web search grounding, reference images, and resolution controls.

Nodespell AI
AI / Image / Google

Google's fast image generation and editing model with fal.ai web search grounding, reference images, and resolution controls.

Model Overview

Nano Banana 2 generates and edits images from natural language prompts. It can use optional reference images and web search grounding when the prompt needs current information.

Best At

  • Prompt-driven image generation and image editing.
  • Using reference images for product, character, or style guidance.
  • Producing context-aware images when web search grounding is enabled.
  • Fast visual iteration with explicit aspect ratio, resolution, and output format controls.

Limitations / Not Good At

  • Web search adds cost and should be used only when current information is needed.
  • Highly constrained layouts, exact typography, and brand-perfect edits may need iteration.
  • Reference image quality and prompt specificity strongly affect edit quality.

Ideal Use Cases

  • Current-event visuals, editorial images, and topical creative assets.
  • Product or character variations from one or more reference images.
  • Social, marketing, and concept imagery that needs fast iteration.
  • Image edits described in natural language.

Input & Output Format

  • Input: Required prompt text, optional image references, aspect ratio, resolution, output format, safety tolerance, web search, and seed.
  • Output: Generated or edited image assets returned on response.

Performance Notes

  • 1K is the default resolution. 2K and 4K increase cost. Web search adds a separate surcharge.

Model Examples (5)

Example Index01 / 05
Example 01

Theater rehearsal

A realistic wide still of a small theater rehearsal.

Open
Source Inputs01
Prompt

Photorealistic wide still of a small community theater rehearsal, actors in everyday clothes blocking a quiet scene on a bare stage, director notes on folding chairs, warm work lights, natural candid production atmosphere.

Parameters07
Prompt
Photorealistic wide still of a small community theater rehearsal, actors in everyday clothes blocking a quiet scene on a bare stage, director notes on folding chairs, warm work lights, natural candid production atmosphere.
Aspect Ratio
16:9
Resolution
1K
Output Format
png
Safety Tolerance
4
Enable Web Search
false
Seed
-1
realismfilmrehearsalstory-reference
Response
Theater rehearsal output
Inputs (2)

Prompt

String

A text description of the image you want to generate or edit.

Multi InputMin: 0Max: 100

Image Input

String

Optional image references for editing or image-guided generation.

Multi InputMin: 0Max: 100
Parameters (7)

Prompt

String

A text description of the image you want to generate or edit.

Required
Default:

Aspect Ratio

String

Aspect ratio of the generated image. Auto lets the model decide from the prompt.

Default: auto

Resolution

String

Resolution of the generated image. Higher resolutions cost more.

Default: 1K

Output Format

String

Format of the output image.

Default: png

Safety Tolerance

String

Content moderation tolerance. 1 is strictest and 6 is most permissive.

Default: 4

Web Search

Boolean

Use web search grounding when the prompt needs current information.

Default: false

Seed

Number

Optional seed for repeatable results. Use -1 to omit.

Outputs (1)

Output

Inferred

Output

Nodespell Team

Creator profile

Type

Node

Status

Official

Package

Nodespell AI

Category

AI / Image / Google

Input

ImageText

Output

Image

Keywords

Image GenerationImage EditAspect ControlResolution ControlSearch Retrieval
Use in Workflow