ComfyUI adds Google's Gemini Nano Banana 2.1 image model via Partner Nodes

AIGoogle's Nano Banana 2.1 is rolling out in Gemini, and a side-by-side test against ChatGPT's Images 2.5 found the results close. Nano Banana 2.1 produced a more convincing greenhouse-in-a-snow-globe scene, rendered a requested travel poster's text and price correctly, and kept a fox consistent across a four-panel comic.

AIGoogle's Gemini Nano Banana 2.1 image generation and editing model is now available in ComfyUI through Partner Nodes, the successor to Nano Banana 2. The model accepts a prompt and up to 14 reference images, outputs at up to 4K, and offers Minimal, Medium and High thinking levels. It adds a 9:21 aspect ratio and, according to the post, costs less per run than Nano Banana 2.