langchaingo
langchaingo copied to clipboard
openaillm: remove internal client, use go-openai client instead
For issue #119 , use go-openai client instead, which simplified a lot of code.
PR Checklist
- [x] Read the Contributing documentation.
- [x] Read the Code of conduct documentation.
- [x] Name your Pull Request title clearly, concisely, and prefixed with the name of the primarily affected package you changed according to Good commit messages (such as
memory: add interfaces for X, Y
orutil: add whizzbang helpers
). - [x] Check that there isn't already a PR that solves the problem the same way to avoid creating a duplicate.
- [x] Provide a description in this PR that addresses what the PR is solving, or reference the issue that it solves (e.g.
Fixes #123
). - [x] Describes the source of new concepts.
- [x] References existing implementations as appropriate.
- [x] Contains test coverage for new functions.
- [x] Passes all
golangci-lint
checks.
@tmc Can this PR be merged? There may be more code conflicts in the future if we wait too long.
This pull request was submitted a long time ago and has not been reviewed, so there may be some conflicts. It seems that it is difficult to fix them now.
This pull request was submitted a long time ago and has not been reviewed, so there may be some conflicts. It seems that it is difficult to fix them now.
Any plans on resolving the conflicts or this pr is dead? I'd love to see github.com/sashabaranov/go-openai usage
Hey sorry all, I let this get stale. I'm a little concerned about taking on the dependency but I'm supportive. can someone take a crack at a rebase?
I will try later.
@bdqfork anything I can do to help here, feel free to reach out on discord or otherwise. Would love to see this go in.
@tmc @bdqfork would use this if merged given "tools" is the new way #507
Is this effort no longer being prioritized?
I didn't choose to close the PR, it needed to be rebased, I'm still open to this if someone would like to contribute it.
@bdqfork Can you make a revision on the latest main branch? Go-openai is indeed a comprehensive support library for openai. A new idea might be to create an openai v2 package, so there would be no conflicts to resolve. If the functionality is rich enough, then the previous openai package can be marked as deprecated.
any thoughts here?
I remain open to this contribution and I welcome a fresh PR.
Can we get a review of the discussion around the official azure sdk for this which might be worth considering on an "officially supported" capacity instead?
The official player is here: https://github.com/openai/openai-go .
I may want to give it a try using https://github.com/openai/openai-go if I have some success with my other PRs