Google Vertex AI - LangWatch

Add the credential

Open Settings > Model Providers, click Add Model Provider and select Vertex AI. Give the row a name, pick the organization, team or project it belongs to, and fill in:

Field	Value
`GOOGLE_APPLICATION_CREDENTIALS`	The service account key as JSON, pasted in full.
`VERTEXAI_PROJECT`	The Google Cloud project id.
`VERTEXAI_LOCATION`	The location, for example `us-central1` or `europe-west4`.

Save the row. Every virtual key whose scope can see the row can now dispatch to it, unless the key restricts its providers. See Virtual keys.

The Vertex AI model provider drawer with GOOGLE_APPLICATION_CREDENTIALS filled in, VERTEXAI_PROJECT acme-ai, VERTEXAI_LOCATION us-central1 and the routing handle vertex_ai — The Vertex AI row: the service account JSON, VERTEXAI_PROJECT and VERTEXAI_LOCATION, all three required. The routing handle defaults to vertex_ai.

Model names

Write vertex/<model>, for example vertex/gemini-2.5-flash. vertex_ai/ is accepted as an alias. The gateway does not infer Vertex AI from a bare name. Add the models you use under Custom Models on the row.

Send a request

curl https://gateway.langwatch.ai/v1/chat/completions \
  -H "Authorization: Bearer $LANGWATCH_VIRTUAL_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "vertex/gemini-2.5-flash",
    "messages": [{"role": "user", "content": "Say hi in one word"}]
  }'

Endpoints

Endpoint	Served
`POST /v1/chat/completions`	Yes
`POST /v1/messages`	Yes, translated
`POST /v1/embeddings`	Yes
`POST /v1/images/generations`, `POST /v1/images/edits`	Yes, non-streaming. See Images
`/v1beta/*`	Yes, the Google GenAI SDKs. See Gemini
`GET /v1/models`	Yes, the names under Custom Models

Last modified on September 6, 2026