Rémi Louf
Rémi Louf
@lapp0 I addressed your comments. Please review one last time before we merge.
> I think this would result in a lot of user complaints about validation errors. Unlike local models which outlines supports, OpenAI doesn't guarantee the output will match the specified...
Yes it happens, but we can return a meaningful error when this happens.
Let's use OpenAI's structured output support, which should be easy! We can always raise an exception if the version of the SDK that is installed is < 1.40 and invite...
Since OpenAI uses JSON Schema under the hood we can always directly use the [to_strict_json_schema](https://github.com/openai/openai-python/blob/40f4cdb52a7494472c32e26c70f54bb41bb2bb57/src/openai/lib/_pydantic.py#L12) function to convert the Pydantic model in a JSON Schema that is compatible with the...
Thank you for the PR! It looks like many tests related to these changes are failing…
I guess that's the standard way of doing it.
- [ ] Add direct link to the respective papers - [ ] Clean code examples - [ ] Clean example introductions - [ ] Make sure that the example...
Can you upgrade outlines for 0.0.43 and try again?
I don't think this design is coherent with the rest of the library; we want to avoid kwargs as much as we possibly can. Here I would simply add a...