Expect MP4 unless the page says otherwise
Where a tool names its file, it is usually MP4. OpenAI's Sora, LTX, HeyGen, Synthesia and BytePlus's Seedance API return one, and Wan's is MP4 with H.264. From 6 source pages, read 2026-09-25; no maker pays to be listed.
Two tools name something else: Luma exports 16-bit EXR and VEED names VP9 with alpha.
MP4 is a delivery format, compact and widely playable, and it keeps less than a colourist needs.
BytePlus's reference shows MOV as an option as well, so a request can ask for a different container.
Twenty tools in the register name no format on the pages read, so test a download before planning a grade around it.
| Tool | Hook | What the documentation says |
|---|---|---|
| OpenAI Sora | Formats | MP4 |
| LTX | Formats | An MP4 returned by the API, from a sync or an async endpoint |
| HeyGen | Formats | One request returns an MP4 |
| Synthesia | Formats | The finished video downloaded as an MP4 |
| BytePlus ModelArk | Formats | MP4 output by default, with MOV shown as an option in its examples |
| Wan (Alibaba Cloud) | Formats | MP4 with H.264 encoding |
Inclusion rule. Readings from the register used on this page. Order. The order the text takes them in.
1As the documentation puts it
you can fetch the final MP4 file with GET /videos/{video_id}/content
OpenAI, video generation guide
The API returns an MP4.
LTX, API page
authenticate, send one request, and get back an MP4
HeyGen, developer documentation: quick start
download the video as an .mp4
Synthesia, developer documentation: video API quickstart
Default value: mp4
BytePlus ModelArk, API reference
The video is in MP4 format with H.264 encoding.
Alibaba Cloud, Wan API reference (text to video)
- Retrieve resultsMP4
- Quick startAn MP4 returned by the API, from a sync or an async endpoint
- Page headerOne request returns an MP4
- Download the videoThe finished video downloaded as an MP4
- Request parametersMP4 output by default, with MOV shown as an option in its examples
- ResponseMP4 with H.264 encoding
2A pipeline stage, described
Read against “Expect MP4 unless the page says otherwise”, this is how SceneMixer describes the matching stage of its own pipeline:
Its character becomes the tone of the episode’s own score, which is planned from the episode’s storyboard — usually one to three pieces — and mixed under the voices.
SceneMixer, Remake: How Spoken Lines and Music Are Read From the Clip, and What Ends Up in Your Episode — “Is the source music reused?” (read 2026-10-13)
You can recomposite with different end-card text, and in step-by-step mode you can reshoot individual segments. The finished file can also be downloaded and shared with a link.
SceneMixer, How to Make an Ad from One Photo — “Can I change the ad after it is finished?” (read 2026-10-13)