stable ridge
Hello everyone,
I'd like to do the following with the API, through the Python client:
- Run text completions
- With o1
- With a JSON schema output
- And an attached file
All the examples point to requiring to create an assistant first, no problemo.
But you can't create assistants with o1 yet, despite o1 accepting PDFs as input.
Looking at devtools, the file is uploaded with the file endpoint, then passed as content metadata in the completion call. I can't seem to do that with the client.
Is this just a limitation of the client, or is it a limitation of the API?