ChatGPT is great at scripts and shot lists. With Typeshoot connected it can produce the clips too: it picks a video model, shows the price, and returns the clip in the chat.
Then it can finish the job: trim, reframe 16:9 to 9:16, join clips, add music and voice, and burn in captions.
Made with Typeshoot
Handheld selfie video. He looks into the camera and says, casually and warmly: "I filled this at seven a.m. It's still hot at lunch. That's it, that's the review." Natural kitchen ambience, no music. Then bold word-by-word captions.
Gemini Omni Flash 720p · 104 credits
A 6-second push-in on the Fernhill can by a forest stream, with a calm voiceover "Brewed slow. Made for early mornings. Fernhill Cold Brew.", a soft acoustic music bed at 30% and bold captions.
Gemini Omni Flash 720p + ElevenLabs v4 + Lyria 3.5 · 130 credits
Slow cinematic push-in toward the green FERNHILL cold brew can on the mossy rock. Morning mist drifts across the stream, sunlight flickers through the pines, a drop of condensation runs down the can. The can and its label stay exactly the same. Sound: a gentle forest stream and a few birds, no music, no voice.
Kling 4.0 Omni · from the image on the left · 157 credits
Handheld selfie video: she shakes the bottle once, then talks to the camera with real enthusiasm, saying: "Okay, this thing blends frozen berries in twenty seconds and fits in my gym bag. Obsessed." Natural kitchen sound, slight handheld shake, authentic social ad.
Gemini Omni Flash 1.2 · from the image on the left
More examples, each with its exact prompt
How it works
- Turn on Developer mode in ChatGPT and add the Typeshoot app.
- Describe the clip or attach a start image.
- Approve the model and the credit price.
- Ask ChatGPT to edit the result: trim, reframe, captions, music.
Try these prompts
PromptMake a 6-second vertical clip of coffee being poured in slow motion, with the sound. Draft it at 360p first and tell me the credits.
PromptAnimate this product photo [attach]: slow push-in, 5 seconds, then add bold captions saying “Hot at lunch”.
Models and prices
| Model | Typical job | Credits | Per $10 pack |
|---|---|---|---|
| Gemini Omni Flash (default) | a 5-second clip at 720p | 85 | 11 |
| MiniMax H3 Max Fast | a 5-second clip at 768p | 40 | 25 |
| Kling 4.0 Omni | a 5-second clip at 720p | 157 | 6 |
| Seedance 2.5 | a 5-second clip at 720p (estimate) | 473 | 2 |
$10 = 1,000 credits. Your assistant shows the price before anything runs, and failed jobs are refunded automatically.
When to use it
- Social clips, product videos and ads planned and made in one ChatGPT conversation.
- People who already work in ChatGPT and don’t want another video app.
- Long films with many shots take planning; draft cheaply first.
Frequently asked questions
Can ChatGPT make videos?
With an app like Typeshoot, yes: ChatGPT sends your brief to a video model, waits for the clip and shows it in the chat. On its own, ChatGPT is focused on text and images.
How much does a ChatGPT video cost?
A 5-second clip is 30 credits at 360p or 85 at 720p on the default model; premium models cost more. See the AI video cost calculator.
Keep going
Try it in your own chat
Typeshoot adds image, video, voice and editing tools to Claude, ChatGPT and other assistants. 100 free credits to start, no card needed.
Get 100 free credits