alisha-1000 commented on issue #40179: URL: https://github.com/apache/beam/issues/40179#issuecomment-6878181607
@damccorm Quick update after looking into this. I found your draft PR #39074, which already implements local model deployment for the ADK handler, so I won't propose a separate design. While exploring, I ran the current handler on master with a local model (LiteLlm + Ollama, llama3.2) on the DirectRunner. It works without changes, including a simple tool call. Along the way I found a bug in session reuse, reported as #41299 with a fix in #41339. The same line is present in #39074. If it would help, I'm happy to test #39074 with local models or help with tests/docs there. -- 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]
