BerriAI/litellm · error · ValueError

vertex_project and vertex_location are required for Vertex A

Error message

vertex_project and vertex_location are required for Vertex AI

What it means

Vertex AI Imagen image edit calls the predict API, whose URL embeds both the GCP project and the region: {base}/v1/projects/{project}/locations/{location}/publishers/google/models/{model}:predict. The config resolves them from vertex_ai_project/vertex_ai_location params, VERTEXAI_PROJECT/VERTEXAI_LOCATION (or VERTEX_LOCATION) env vars, module-level litellm.vertex_project/vertex_location, and secrets; if either is missing this ValueError fires. Important nuance: unlike the Gemini handlers, api_base does NOT bypass this check — project/location are needed to build the path even when api_base is supplied.

Source

Thrown at litellm/llms/vertex_ai/image_edit/vertex_imagen_transformation.py:132

            project_id=vertex_project,
            custom_llm_provider="vertex_ai",
        )
        return self.set_headers(access_token, headers)

    def get_complete_url(
        self,
        model: str,
        api_base: str | None,
        litellm_params: dict,
    ) -> str:
        """
        Get the complete URL for Vertex AI Imagen predict API
        """
        vertex_project: Final = self.safe_get_vertex_ai_project(litellm_params) or self._resolve_vertex_project()
        vertex_location: Final = self.safe_get_vertex_ai_location(litellm_params) or self._resolve_vertex_location()

        if not vertex_project or not vertex_location:
            raise ValueError("vertex_project and vertex_location are required for Vertex AI")

        # Use the model name as provided, handling vertex_ai prefix
        model_name = model
        if model.startswith("vertex_ai/"):
            model_name = model.replace("vertex_ai/", "")

        if api_base:
            base_url = api_base.rstrip("/")
        else:
            base_url = get_vertex_base_url(vertex_location)

        return f"{base_url}/v1/projects/{vertex_project}/locations/{vertex_location}/publishers/google/models/{model_name}:predict"

    def transform_image_edit_request(
        self,
        model: str,
        prompt: str | None,
        image: FileTypes | None,

View on GitHub (pinned to 77b7c6c40c)

Solutions

1. Pass vertex_ai_project='my-project' and vertex_ai_location='us-central1' on every image_edit call targeting vertex_ai/imagen-* edit models
2. Or export VERTEXAI_PROJECT and VERTEXAI_LOCATION in the process environment
3. Or set litellm.vertex_project / litellm.vertex_location module-level defaults
4. Do not rely on api_base alone for Imagen edit — it cannot substitute for project/location

Example fix

# before (raises even with api_base)
resp = litellm.image_edit(
    model='vertex_ai/imagen-3.0-capability-001',
    prompt='remove the background',
    image=open('cat.png', 'rb').read(),
    api_base='https://my-proxy.example.com',
)

# after
resp = litellm.image_edit(
    model='vertex_ai/imagen-3.0-capability-001',
    prompt='remove the background',
    image=open('cat.png', 'rb').read(),
    vertex_ai_project='my-gcp-project',
    vertex_ai_location='us-central1',
    api_base='https://my-proxy.example.com',
)

import os

project = os.environ.get('VERTEXAI_PROJECT')
location = os.environ.get('VERTEXAI_LOCATION') or os.environ.get('VERTEX_LOCATION')
if not (project and location):
    raise RuntimeError('imagen edit needs project+location even when api_base is set')

try:
    resp = litellm.image_edit(model='vertex_ai/imagen-3.0-capability-001', prompt=p, image=img)
except ValueError as e:
    if 'vertex_project and vertex_location are required' in str(e):
        raise RuntimeError('configure vertex_ai_project/vertex_ai_location (api_base will not bypass this)') from e
    raise

When it happens

Trigger: litellm.image_edit(model='vertex_ai/imagen-3.0-capability-001', prompt=..., image=f, api_base='https://my-proxy') still raises, because the project/location check precedes the api_base branch; likewise any call without params or env vars set.

Common situations: Assuming a proxy api_base removes the need for project/location (it does not for Imagen edit); credentials-only setup (GOOGLE_APPLICATION_CREDENTIALS) with no project/location; env vars missing in dockerized proxy deployments.

Understand the failure class

Background: "X is required", "must be set", "cannot be empty": the missing-required-config error family, from Vertex AI project/location to WeChat keys — this error's family across 18 libraries.

Related errors

• vertex_project and vertex_location are required for Vertex A
• vertex_project and vertex_location are required for Vertex A
• vertex_project and vertex_location are required for Vertex A
• Vertex AI Imagen image edit requires at least one reference
• Vertex AI Imagen image edit requires a prompt.

AI-assisted analysis of BerriAI/litellm@77b7c6c40c (2026-08-18). Data as JSON: /api/errors/d4d804df833ae559. Report an issue: GitHub.