Documentation outdated and Openrouter and retries are broken not receiving instruction when invalidation

[philmas]

• [x ] This is actually a bug report.
• I am not getting good LLM Results
• I have tried asking for help in the community on discord or discussions and have not received a response.
• I have tried searching the documentation and have not found an answer.

What Model are you using?

• [ x] gpt-3.5-turbo
• [ x] gpt-4-turbo
• [x ] gpt-4
• [ x] Other (please specify)

Describe the bug
A clear and concise description of what the bug is.

Cannot seem to get Openrouter to work nor is it possible to get the self-correction to work.

To Reproduce
Steps to reproduce the behavior, including code snippets of the model and the input data and openai response.

Follow your own documentation :) self-correction to work. It wont work.

Expected behavior
A clear and concise description of what you expected to happen.

Screenshots
If applicable, add screenshots to help explain your problem.

[jxnl]

@cursoragent please take a look and solve this. If you need repro/versions/scope, ask here.

[cursor]

Thanks — I’m on it. I’m going to:

• Reproduce the OpenRouter setup using the current docs (model + base_url, headers, and client patching).
• Reproduce the “self-correction / self_critique” example and verify the retry flow is actually sending the validation error + instructions back to the model.

If I can’t reproduce cleanly, I’ll ask for a minimal repro. The most useful details would be:

• instructor version, openai version, and Python version
• The exact OpenRouter config you’re using (base_url, model name, and how you pass the API key)
• A short code snippet + the raw error/traceback (or a redacted request/response)

I’ll report back with either a fix + doc update, or a concrete repro request.