It’s been a while but Google has a new image model - Nano Banana 2.1 (gemini-nano-banana-2.1)
It is their next-generation multimodal reasoning model.
- It can take text and image input with a 1M token context window
- It can output images and text with a 64K context window
The model was released in October so, if you are reading this, the model is available now.
They have it in all of the expected places
- Gemini App
- Google AI Studio
- Gemini API
- Google Search
- Google Flow
- Google Stitch
- Etc.
I do appreciate that Google publishes the known limitations with their models.
Some of the issues with this model are -
- Poor text rendering in small text, long paragraphs, page length
- Character consistency issues
- Limitations in advanced capabilities with world knowledge, 3D reasoning, and factuality
Watch a Video:
.png?width=600&height=496&name=Next-Level%20AI%20Workshops%20(2000%20x%201000%20px).png)
From "Cool Image" to Real Workflow
Image AI is about more than making pretty pictures. Generate, edit, and automate visual content in our hands-on AI workshops.
Comments