zqr10159 commented on PR #4392: URL: https://github.com/apache/hertzbeat/pull/4392#issuecomment-5908333957
Thanks for the fix and the follow-up request capture. The configurable value and compatibility default look reasonable on source review. A few points before merging: 1. Please make the warning in LlmConfig.buildChatOptions identify only the fields that actually fall back. With an unknown code, a configured baseUrl and a null model, it says the OpenAI endpoint and model are used, although the custom endpoint is retained. Please cover both partially configured cases. 2. Please reword the PR description sentence "This PR does not fix #3971". GitHub currently lists #3971 under issues that this PR will close, despite the explicit scope disclaimer. Wording such as "The problem reported in #3971 is outside this PR's scope" should avoid the closing-keyword match; please verify the Development section afterward. 3. Backend and license workflows are still awaiting maintainer approval, so their results are not available yet. Thanks for distinguishing the direct Spring AI capture from a full application run. I have not run the application locally. The API-only temperature setting and the retained 0.3 default are understood as the scope here, rather than a fix for providers that require temperature to be omitted. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
