Langchain prioritizes OpenAI for examples and documentation - all examples should have open-source model version

[matsuobasho]

Checked

• I searched existing ideas and did not find a similar one
• I added a very descriptive title
• I've clearly described the feature request and motivation for it

Feature request

I've noticed that most examples in the LangChain docs are based on using OpenAI keys. I know there are sections on using open-source models, but those are rather the exception.

Motivation

I respect that LangChain is an open-source effort which goes where the community takes it. However, I don't want to pay OpenAI for tokens. I've experimented with using open-source lms and it generally works but it feels like it often requires a hack.

Is this a conscious decision (perhaps even supported financially by OpenAI) or was it just a decision of convenience?

I'm open to having my opinion changed or being educated on the matter.

For a small example of what I mean, take a look at the issue I opened a few days ago that's still leaving me very confused about how to proceed:
#18979

Proposal (If applicable)

If Langchain really wants to play an important part in the agent ecosystem for accessibility to all, all documentation has to have an open-source version by default. I'm happy to do my small share to update documentation.