Is Nano Banana best for creating 3D models? We tested it against ChatGPT, Grok and Qwen

Gemini’s 2.5 Flash image model, also known as Nano Banana, has become a social media darling in the last few days. Most popular apps like Twitter/X, Instagram, Facebook, and LinkedIn are filled with 3D model images generated using the latest AI model from Google. While Gemini’s Nano Banana is in a league of its own due to its ability to create complex, high-quality, photo-realistic images faster than most other image generation models, the competition in the AI space is intense.

Nano Banana isn’t the only AI image generation model that can create similar 3D figurines, even for free users. Here’s a look at the top Nano Banana alternatives you can find.

Top Nano Banana alternatives you can check out:

To check the quality of various image generation models compared to Nano Banana, we gave the top three free alternatives—Qwen Image Edit, ChatGPT (GPT-5), and Grok AI—the same image and prompt to generate a 3D model of the user.

Prompt used for this experiment:

“Create a 1/7 scale commercialized figurine of the characters in the picture, in a realistic style, in a real environment. The figurine is placed on a computer desk. The figurine has a round transparent acrylic base, with no text on the base. The content on the computer screen is a 3D modeling process of this figurine. Next to the computer screen is a toy packaging box, designed in a style reminiscent of high-quality collectible figures, printed with original artwork. The packaging features two-dimensional flat illustrations.”

For reference, here’s the image that Nano Banana generated using a picture I uploaded.

Qwen:

In case this is the first time you are hearing about Qwen, it’s a free-to-use chatbot developed by Alibaba Group. Qwen’s latest image generation model, Qwen-Image-Edit, was launched by the company last month and garnered a lot of praise on social media for its ability to make complex and precise image modifications using natural language prompts.

Qwen-Image-Edit can understand the conceptual meaning of an image and is capable of adding pixel-level details like textures, colors, and lighting while ensuring that the changes to the image look realistic and high-fidelity.

Here’s the image we managed to generate with Qwen. In my experience, it felt like Qwen was able to get the surroundings mostly right, and the image quality and sharpness looked much better than that of Gemini, but the model could not get the facial features quite right.

ChatGPT:

OpenAI’s GPT-5 is its best model yet and comes with the ability to generate and edit images. Its predecessor, GPT-4o, had won a lot of plaudits earlier this year when it received a native image generation update that allowed it to not only transform users’ pictures into Ghibli-style art but also into real-life action figures and more.

Despite competition from Gemini and Qwen, ChatGPT continues to maintain its position. Personally, I feel that GPT-5 is slightly better at understanding instructions than Gemini or Qwen, but where it lacks is the slow speed of image generation and not getting the facial features right. ChatGPT also only allows creating just two images per day, which is very low compared to Nano Banana, Qwen, and even Grok.

Here’s the image we managd to generate using ChatGPT.

Grok:

xAI’s Grok AI used to be among the top free image generators until Nano Banana, Qwen, and ChatGPT came into the picture. The model is still good for generating random images, but for making complex edits to images and creating photo-realistic 3D models, it may not be the right choice.

Here’s the image we managed to generate using Grok AI.

One thing that does work in Grok’s favor, though, is that it is currently among the only BIG Tech-offered AI chatbots that can generate videos for free. So, if you have already generated a 3D model using Gemini, ChatGPT, or Qwen, Grok could easily turn it into a small video with accompanying sounds.