For the Conversation Images feature to work, we need to fine-tune a specialized AI model to teach it to generate similar images than your Chatbot Image. That process takes some time (about 7 minutes per bot), therefore we've created a training queue. All our I'm All In subscribers can now request the training of any untrained chatbot so that it can be available for Conversation Images. This includes training your private bots, or any public bot that has not been trained yet. To request training, you just need to click on the greyed out Image icon on the Chat Input:

For the moment, we've set the limit to a maximum of 4 training requests per day, however that limit will likely evolve based on the volume of training requests.

Some of the future improvements to Conversation Images we want to focus on in the near future:

• Adding more than one image per bot to allow better fine-tuning
• Editing the Image Generation Prompt to have more control of the generated image
• Training a custom LLM to generate better image prompt
• Using a different diffusion model that can generate more complex scenery.

All this to say that this feature will continue to evolve!

A side note to this, better text generation continue to be our main priority, and we plan to be sharing our progress on this with you very soon :)