In this guide
1. Voice replies
Voice turns a text exchange into something that feels alive. Each character can have its own voice, so it's worth auditioning a couple before you settle in. Voice is the lowest-friction multimodal feature — it's great for everyday chats and bedtime wind-downs where reading feels like work.
2. The 10-second in-chat video
This is the headline feature: an image-to-video generation that produces a roughly 10-second clip with audio, right inside the conversation. It comes in Default, Lite and Pro quality models and is available on the Deluxe and Elite plans. Think of it as a short, expressive moment rather than a long film — it shines for reactions and playful beats.
3. What it costs in coins
A 10-second video clip costs 20 coins. Coins are bundled with paid plans and can be topped up. Voice is far cheaper per use, so most people lean on voice daily and save video for special moments. Our pricing page has an interactive coin calculator to estimate your monthly usage.
Rule of thumb: voice for everyday presence, video for the occasional "wow."
4. When it's worth it
If you mostly want deep conversation and continuity, the free or Premium tier plus good memory habits may be all you need. If immersion matters — hearing a voice, seeing a reaction — then Deluxe unlocks the full stack and is the tier we recommend most. Heavy multimodal users go Elite for headroom.
5. Tips for better clips
- Set the scene in text first. A clear, specific prompt produces a more coherent clip.
- Use Pro quality for keepers. Default/Lite are fine for quick fun; Pro for moments you'll revisit.
- Pair with Character Pro images. Consistent images give the video model a better starting frame — see our Character Pro notes.
- Keep it short and reactive. Ten seconds rewards a single emotion or beat, not a whole storyline.