Free OmniGen Online

OmniGen is a versatile multi-modal image generation model designed to simplify the creation and modification of images through text prompts. Unlike traditional models, it eliminates the need for loading additional network modules, such as ControlNet and IP-Adapter, and bypasses complex preprocessing steps like face detection and pose estimation. This streamlined approach allows users, whether beginners or professionals, to generate high-quality images effortlessly.

Quick Start of OmniGen

How to Use OmniGen to Generate and Refine Images？

1. Add OmniGen to Your MimicPC Dashboard: Start by logging into your MimicPC account and adding OmniGen to your dashboard. This will allow you to access its powerful image generation features without installation.
2. Choose Your Hardware: Select the appropriate hardware configuration. Make sure to choose a setup with large-pro or higher VRAM to ensure optimal performance. Once you’ve made your selection, click on "Create & Start."
3. Input Your Prompt and Adjust Settings: In the OmniGen interface, input your desired text prompt. You can also customize the settings according to your specific needs to tailor the output.
4. Generate Your Image: Click on "Generate" to create the image based on your prompt. Once the generation is complete, you can download the image directly to your device.
5. Refine the Generated Image: If you wish to make adjustments to the generated image, upload it and provide a prompt detailing the changes. For example, you can use the prompt: "<img><|image_1|></img> Remove the woman's surfboard." Then click "Generate" to see the refined result.
6. Automatic Object Identification: You will notice that the surfboard has been removed from the image. OmniGen’s advanced feature recognition allows it to identify objects automatically and implement changes based on your prompts.
7. Create Images from Multiple Inputs: You can also input 2 to 3 photos along with a text prompt to generate a new image. For instance, you could use the prompt: "Two men are drinking in a bar. One man is <img><|image_1|></img>. Another man is <img><|image_2|></img>." Ensure that all input images are of the same size for optimal results.