Repository navigation
GitHub Actions workflow discussion_answering.yml fails on Q&A discussion creation with a Vertex AI 404 for gemini-3.5-flash #6104
Description
Activity
Hi @ftnext , Thanks for flagging this model routing issue — the availability gap for the hardcoded
gemini-3.5-flashunder regional endpoints is clearly confirmed, and running the workflow without targeting a global endpoint immediately fires the 404 with no other setup needed.The approach in PR #6113 (making the discussion-answering model configurable rather than hardcoding it, while keeping the default configuration flexible) looks like exactly the right fix. This allows users to either opt into widely supported models that exist in standard regional endpoints, or supply their own custom model parameters.
Could you also confirm the configuration override holds for your local runner environment — specifically that you can successfully specify an alternate model (or apply the
"global"location override to keep usinggemini-3.5-flash) — and verify that this setup successfully unblocks execution on your end?I tested this locally with a Vertex AI-enabled GCP project.
Confirmed:
- The
LLM_MODEL_NAMEoverride is applied to the discussion answering agent. - With
LLM_MODEL_NAME=gemini-3.5-flashandGOOGLE_CLOUD_LOCATION=global, the actual answering agent loads withroot_agent.model == "gemini-3.5-flash".
The PR's default model of
gemini-2.5-flashmay not be safe for this agent's current mixed tool configuration.400 INVALID_ARGUMENT: Multiple tools are supported only when they are all search tools.
- The
- added a commit that references this issue
on Jul 13, 2026 - added a commit that references this issue
on Aug 11, 2026
Run: https://github.com/google/adk-python/actions/runs/27451592905
Discussion: #6103
Error:
The failing step is
Run Answering ScriptinADK Answering Agent for Discussions. The workflow setsGOOGLE_GENAI_USE_VERTEXAI=1, and the model is hardcoded in:contributing/samples/adk_team/adk_answering_agent/agent.pycontributing/samples/adk_team/adk_answering_agent/gemini_assistant/agent.pyThis looks similar to #5696, but that issue covered triage workflows using Gemini Developer API. This failure is in the discussion answering workflow using Vertex AI.
Possible cause:
GOOGLE_CLOUD_LOCATIONmay point to a region wheregemini-3.5-flashis unavailable, or the Actions GCP project/service account does not have access to that model. Current docs listgemini-3.5-flashavailability forglobal,us, andeu.