> For the complete documentation index, see [llms.txt](https://xfigura.gitbook.io/xfigura-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://xfigura.gitbook.io/xfigura-docs/models/image-models/nano-banana-2.1.md).

# Nano Banana 2.1

Image generation and conversational editing with Google's Nano Banana 2.1.

**Nano Banana 2.1** is Google's update to [Nano Banana 2](/xfigura-docs/models/image-models/nano-banana-2.md) for image generation and conversational editing. Google reports improvements in realism, prompt following, text rendering, and consistency across successive edits while retaining the Flash model line's focus on speed and efficiency.

Google's model code is `gemini-nano-banana-2.1`. It is distinct from **Nano Banana 2 Lite** and **Nano Banana Pro**.

### Use it in xFigura

Nano Banana 2.1 is available in **Text to Image** and **Image to Image**. Choose it from the node's model menu, add your prompt and any reference images supported by that workflow, then review the run estimate before generating.

Its credit cost matches Nano Banana 2. Provider features listed below do not imply that every option is exposed in xFigura; use the controls shown in the node.

### Example uses

* Refine a product scene through several rounds of changes
* Develop posters, diagrams, or infographics with specified text and layout
* Combine references while checking that subjects and objects stay recognizable
* Explore wide compositions and higher-resolution image outputs

### Inputs and outputs

| Provider capability      | Supported                                                                |
| ------------------------ | ------------------------------------------------------------------------ |
| Image workflow inputs    | Text prompts and reference images                                        |
| Provider outputs         | Image and text                                                           |
| Provider resolution      | 1K, 2K, and 4K; default 1K                                               |
| Provider reference limit | Up to 14 images; up to 4 characters and 10 objects                       |
| Provider options         | Google Web and Image Search grounding; minimal, medium, or high Thinking |

These are underlying model capabilities from Google's [model documentation](https://ai.google.dev/gemini-api/docs/models/gemini-nano-banana-2.1). Check the xFigura model menu for current availability. Node controls, reference limits, and resolution options depend on the xFigura workflow.

### Strengths

* Improved text rendering and infographic layouts
* Better subject consistency during repeated edits
* Improved realism and fewer tiling artifacts in wide 2K and 4K outputs

### Limitations

* Small text and long passages can remain blurry or inaccurate
* Characters and objects may change between the reference and output
* Masked or doodle-based edits can follow instructions only partially or retain drawn marks
* Left/right placement, spatial reasoning, and factual details can be wrong

Review the result against the brief before using it. Google's [model card](https://deepmind.google/models/model-cards/nano-banana-2-1/) describes these limitations; improved text rendering and consistency do not guarantee exact results.

### Choose within the Nano Banana family

| Model                                                                         | Start here when                                                                                 |
| ----------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------- |
| [Nano Banana 2 Lite](/xfigura-docs/models/image-models/nano-banana-2-lite.md) | You need fast concepting or straightforward edits at 1K                                         |
| [Nano Banana 2](/xfigura-docs/models/image-models/nano-banana-2.md)           | You are continuing an existing Nano Banana 2 workflow                                           |
| **Nano Banana 2.1**                                                           | You need the updated model's text rendering, subject consistency, or image editing capabilities |
| [Nano Banana Pro](/xfigura-docs/models/image-models/nano-banana-pro.md)       | You are continuing a Pro workflow with complex composition or reference constraints             |
| [Nano Banana](/xfigura-docs/models/image-models/nano-banana.md)               | You are continuing a workflow built around the legacy model                                     |

### xFigura tip

State the change and what must remain fixed. For example: “Change the chair's upholstery to dark green velvet. Keep its shape, position, and the room's lighting unchanged.” Check each result before making the next edit. For text, provide the exact copy and placement, then inspect spelling and readability.

### Credits and timing

Nano Banana 2.1 uses the same credit cost as Nano Banana 2 in xFigura. Review the estimate before running a job. Provider API prices and speed claims do not define xFigura credit costs or end-to-end processing time.

### Continue

* [Compare image models](/xfigura-docs/models/image-models.md)
* [Text to Image](/xfigura-docs/generate/image-node/text-to-image.md)
* [Image to Image](/xfigura-docs/generate/image-node/image-to-image.md)