Text to Image (Using AI)

About

Text to Image (Using AI) extension lets you create an image from a text prompt by using artificial intelligence.

To work with this addon, please press the toolbar button to open the interface. Then, wait for the app to load completely. Once the addon is ready, please write a text prompt (at the bottom). When the prompt is ready, press the - Start - button. Depending on your machine, it may take up to a few seconds to minutes for the AI engine to generate the image (at the top section). To download the image please press the download button at the top right corner of the image box. Currently, the final image size is limited to 384x384 pixels.

Please note that this is still a work in progress and might not be able to generate a correct image for all text prompts. Moreover, during the image generation process, you might see some lags or freezes in your browser. So please be patient until the AI engine completes the job. This addon uses a machine-learning library for the web called Transformers.js (https://huggingface.co/docs/transformers.js/). It also uses a text-to-image library called DeepSeek Janus (https://huggingface.co/deepseek-ai/Janus-1.3B).

Note 1: When you first open the extension, it needs to download training data. The complete trained database for this addon is rather large (~2GB), so please be patient while it is being downloaded to your browser. Once the addon is ready, it does not need an internet connection to operate. You can work offline and create as many images as you need without any limitations.

Note 2: This addon is working only if WebGPU (https://www.w3.org/TR/webgpu/) is supported in your browser. The CPU-only version of this AI engine is not currently supported.

If you have a feature request or found a bug to report, please fill out the bug report form on the addon's homepage (https://mybrowseraddon.com/text-to-image-using-ai.html).