Just say it. Cotekumi fills in the rest.
Top image and video models in one conversation. It remembers what you’ve made, so one line is enough.


One line in. The rest, filled in.
Name what you want. Cotekumi brings back the character from earlier, keeps its features, writes the full prompt, picks the model and shows the price.
Elsewhere, you do this every time
- Upload the reference picture again
- Describe its features again
- Write the whole prompt
- Choose a model and its settings
- Find out the cost afterwards
In Cotekumi
Say what you want.
The other five happen on their own. You see what it used, and the price, before you send.
Real conversations, start to finish.
These are the exact messages behind each picture and clip, sent as written.
- 5 messages
- No uploads
- No settings changed
- 210 credits in total
- 3 messages
- No uploads
- No settings changed
- 39 credits in total
Also in Cotekumi
The price before you send
Every picture and video shows its cost first. Nothing is charged by surprise.
The right model, and why
Automatic picks the model that suits the request and tells you why it chose it.
Every version kept
Regenerate or edit as often as you like, and switch back to any earlier version.
Stop before it starts
Stop a picture or video before it reaches its model and nothing is charged.
Made for your phone
Designed on a phone first, not shrunk down from a desktop.
Inside your AI tools
Connect Codex or Claude Code and make pictures and video from there.















