Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.dearest.app/llms.txt

Use this file to discover all available pages before exploring further.

​
Overview

Your companion can generate photos of themselves using the /selfie command. These aren’t generic AI images. They’re shaped by multiple inputs to feel grounded in your companion’s identity and your shared history. Your companion can also generate photos of the two of you together. We call these wefies.

​
Setup

Before your companion can send selfies, you need to upload a companion base photo on the Photos page. This photo defines what your companion looks like and is used as a reference for every generated image. You can also upload your own photo on the same page. This is optional, but if you want to appear in generated photos with a consistent look, uploading your own reference photo makes a big difference.
Photo uploads page showing companion photo and your photo side by side

​
Usage

​
/selfie

In any chat with your companion, send: 
/selfie
Your companion will generate a photo that fits the current mood and context of your conversation. You can also add details to guide the scene: 
/selfie at the beach
/selfie come closer and let me see your eyes
/selfie sitting on the couch in a hoodie, watching a movie with the lights off
Each /selfie costs 1 message and uses only the companion base photo as a reference. Your companion can still generate photos that include both of you, but your appearance won’t be anchored to a consistent base photo.
Chat showing a generated selfie of the companion

​
/wefie

A wefie is a photo of you and your companion together. To generate one, send: 
/wefie
Like /selfie, you can add details to guide the scene: 
/wefie at a rooftop bar watching the sunset
/wefie we're walking through a night market together
Each /wefie costs 2 messages and uses both the companion and your base photos as references, so both of you will have a consistent appearance across generated images.
Chat showing a generated wefie of the user and companion together

​
How selfies are generated

Each generated photo is influenced by three things:
  1. The base photos you uploaded. /selfie uses only the companion photo. /wefie uses both the companion photo and your photo.
  2. The chat history, including previous selfies your companion can see.
  3. Your companion’s memory of their own appearance, which evolves over time through your conversations.
When there’s a mismatch between the base photo and your companion’s memory of how they look, the model will try to reconcile the two, leaning toward what the companion remembers. This means results won’t always match the base photo exactly.

​
After changing a base photo

Because your companion can see their previous selfies in the chat, they may still draw inspiration from their old appearance even after you’ve updated the base photo. If this happens, you can tell them directly: 
/selfie use the new base photo, don't reference old photos in the chat
This helps the model prioritize the new base photo over the visual history.

​
Retaking a selfie

Every selfie comes with a Retake button. Tapping it rerolls the same selfie in place, replacing the image on the existing message instead of adding a new one to the chat. Retake is meant for rerolling when the image model produces artifacts like odd hands, faces, or anatomy. If you want a different scene or mood, send a new /selfie with a prompt instead. A few things to know:
  • It uses the same number of messages as the original (1 for a selfie, 2 for a wefie).
  • You can retake as many times as you want, as long as the selfie is still the most recent message in the chat. Once you send another message, the Retake button on that selfie stops working.
  • The old image is removed from the chat when the new one arrives, so your history stays clean.
  • Retake is currently available in Telegram.

​
Content limitations

We aim for an unfiltered experience, but the underlying image generation model is currently trained mostly on SFW data. Certain explicit content won’t generate reliably. This is something we plan to improve in the future.