I have some code that hits the Vertex AI endpoint but need to keep it region specific.
I am using the London region europe-west2 but get a 404 error when trying to use gemini-2.5-flash. If I use an older model gemini-1.5-flash-002 the code works so this is an availability issue with the API rather than an issue my end.
Given the data residency information here: https://cloud.google.com/vertex-ai/generative-ai/docs/learn/data-residency
It appears that there is a gemini-2.5-flash model available in the region, even if it is only 128k.
Initially I thought there might be a slightly different name depending on the 128k vs 1M but this doesn’t appear to be the case as gemini-1.5-flash-002 has two different versions that again depends on region, but this is working.
Why am I getting this error?
Thank you for your help.
Hi pyrexia,
Welcome to the Google Cloud Community!
The 404 error is likely occurring because the gemini-2.5-flash model is not yet fully available in the europe-west2 region. To work around this, wait for the rollout to complete in that region, ensure your client SDK is updated to the latest version, and, if immediate access is required, use gemini-1.5-flash-002 in europe-west2 or temporarily route requests to gemini-2.5-flash in us-central1, as long as this complies with your data residency requirements.
Additionally, Keep an eye on Vertex AI release notes. They frequently announce when new models or features become generally available in additional regions.
Was this helpful? If so, please accept this answer as “Solution”. If you need additional assistance, reply here within 2 business days and I’ll be happy to help.