GPT-Video Full

The official libraries for the GPT-Video API. GPT-Video turns a long video into short vertical clips with captions and face framing: your program uploads a video, GPT-Video finds the moments that work as clips, and your program exports the ones it keeps as MP4 files.

js, python, n8n and postman each have a README that takes you to a first call, and a CHANGELOG.

How they behave

The JavaScript SDK, the Python SDK and the n8n nodes share one set of rules, written down as test vectors in vectors and run by each of them. The Postman collection makes its Idempotency-Keys in the format of the same vectors.

• A call that spends sends one Idempotency-Key, made before the first attempt and sent again on every retry, so a retry never pays twice. When such a call ends without a clear answer, its error says so and carries the key, and calling again with that key returns the answer of the first call if it succeeded, without paying again.
• Each attempt has a timeout and each call a deadline. A request is sent again only when that is safe, and a request that carries the API key never follows a redirect, because the redirect could take the key to another address.
• A webhook is checked on the raw bytes of the request, in constant time.
• The API key never appears in an error or in a printed client.

Files whose name contains .generated. or ends in _generated.py are made from the API reference and replaced whole at every release, so a change to one of them is lost. The API reference is at https://www.gpt-video.com/developers/api.

Help

Questions and security reports: [email protected]. Bugs: open an issue in this repository.

MIT license. See LICENSE.