Vertex AI SDK (Python)

1.142.0 verified Tue May 12 auth: no python install: verified quickstart: reviewed

Google Cloud Vertex AI SDK for Python. Two separate PyPI packages exist: google-cloud-aiplatform (GAPIC clients, full Vertex AI platform) and vertexai (higher-level SDK, same underlying package). Install is google-cloud-aiplatform, import is vertexai — classic name mismatch. Generative AI modules (vertexai.generative_models, vertexai.language_models, etc.) deprecated June 24, 2025 and will be removed June 24, 2026. Current version: google-cloud-aiplatform 1.142.0 / vertexai 1.71.1.

pip install google-cloud-aiplatform

Common errors

error ModuleNotFoundError: No module named 'vertexai' ↓

cause The `vertexai` SDK is part of the `google-cloud-aiplatform` package, which was not installed or installed incorrectly.

fix

pip install google-cloud-aiplatform

error google.api_core.exceptions.InvalidArgument: 400 The model specified does not exist in this region. ↓

cause The specified model name (e.g., `gemini-pro`) is either misspelled or not available in the Google Cloud region specified during `vertexai.init()` or model instantiation.

fix

Verify the model ID and ensure the location parameter in vertexai.init() or GenerativeModel() call matches a region where the model is available.

error google.auth.exceptions.DefaultCredentialsError: Could not automatically determine credentials. ↓

cause The environment lacks valid Google Cloud credentials, or the service account/user running the code does not have the necessary permissions to access Vertex AI resources.

fix

Authenticate using gcloud auth application-default login, set the GOOGLE_APPLICATION_CREDENTIALS environment variable, or ensure a service account with the 'Vertex AI User' role is active.

error ValueError: Project ID must be specified ↓

cause The `vertexai.init()` function could not automatically determine the Google Cloud project ID from environment variables, and it was not explicitly provided.

fix

Call vertexai.init(project="your-project-id", location="your-region") or set the GCLOUD_PROJECT environment variable.

error AttributeError: 'NoneType' object has no attribute 'generate_content' ↓

cause The `GenerativeModel` or `ChatSession` object was not successfully initialized, often due to an issue during its creation (e.g., invalid model name, region, or authentication failure), resulting in the variable being `None`.

fix

Ensure vertexai.init() is called with correct project/location, the GenerativeModel is instantiated with a valid model name, and verify that the model object is not None before attempting to call generation methods.

Warnings

breaking Install package is google-cloud-aiplatform but import is 'import vertexai'. LLMs consistently confuse this and generate 'from google.cloud import vertexai' or 'pip install vertexai-sdk'. Both are wrong. ↓

fix pip install google-cloud-aiplatform; import vertexai

breaking vertexai.generative_models, vertexai.language_models, vertexai.vision_models, vertexai.tuning, vertexai.caching deprecated June 24, 2025. Will be removed June 24, 2026. ↓

fix Migrate to google-genai with vertexai=True. See migration guide at cloud.google.com/vertex-ai/generative-ai/docs/deprecations/genai-vertexai-sdk

breaking Vertex AI does not use API keys. Uses Application Default Credentials (ADC). No api_key parameter. LLMs frequently hallucinate an api_key argument on vertexai.init(). ↓

fix Run: gcloud auth application-default login. Or set GOOGLE_APPLICATION_CREDENTIALS to service account JSON path.

gotcha Two PyPI packages: google-cloud-aiplatform and vertexai. They install the same underlying package. Installing either gives you both namespaces. Pinning both to different versions causes conflicts. ↓

fix Pin only one. Prefer google-cloud-aiplatform for consistency with GCP tooling.

gotcha vertexai.init() must be called before any Vertex AI SDK calls. Forgetting this causes authentication or project resolution errors. ↓

fix Always call vertexai.init(project='...', location='...') at startup.

gotcha location parameter is required and must be a valid GCP region. 'us-central1' is most common. Wrong or missing location raises InvalidArgument errors. ↓

fix Always pass location='us-central1' or your preferred region.

Install

pip install vertexai

pip install 'google-cloud-aiplatform[agent_engines,adk]'

Install compatibility verified last tested: 2026-05-12

python os / libc variant status wheel install import disk

3.10 alpine (musl) agent_engines,adk - - 18.67s 843.9M

3.10 alpine (musl) google-cloud-aiplatform - - 16.93s 323.8M

3.10 alpine (musl) vertexai - - 13.26s 272.5M

3.10 slim (glibc) agent_engines,adk - - 6.51s 882M

3.10 slim (glibc) google-cloud-aiplatform - - 5.09s 389M

3.10 slim (glibc) vertexai - - 4.09s 262M

3.11 alpine (musl) agent_engines,adk - - 21.51s 905.9M

3.11 alpine (musl) google-cloud-aiplatform - - 17.72s 360.2M

3.11 alpine (musl) vertexai - - 13.75s 299.1M

3.11 slim (glibc) agent_engines,adk - - 11.76s 944M

3.11 slim (glibc) google-cloud-aiplatform - - 8.59s 425M

3.11 slim (glibc) vertexai - - 6.50s 288M

3.12 alpine (musl) agent_engines,adk - - 20.93s 889.5M

3.12 alpine (musl) google-cloud-aiplatform - - 16.47s 347.2M

3.12 alpine (musl) vertexai - - 12.87s 285.3M

3.12 slim (glibc) agent_engines,adk - - 13.46s 928M

3.12 slim (glibc) google-cloud-aiplatform - - 9.86s 412M

3.12 slim (glibc) vertexai - - 7.26s 274M

3.13 alpine (musl) agent_engines,adk - - 19.64s 881.6M

3.13 alpine (musl) google-cloud-aiplatform - - 15.98s 341.3M

3.13 alpine (musl) vertexai - - 11.99s 282.7M

3.13 slim (glibc) agent_engines,adk - - 13.66s 922M

3.13 slim (glibc) google-cloud-aiplatform - - 9.66s 408M

3.13 slim (glibc) vertexai - - 7.04s 272M

3.9 alpine (musl) agent_engines,adk - - 16.41s 589.8M

3.9 alpine (musl) google-cloud-aiplatform - - 15.73s 319.3M

3.9 alpine (musl) vertexai - - - -

3.9 slim (glibc) agent_engines,adk - - 6.08s 656M

3.9 slim (glibc) google-cloud-aiplatform - - 5.46s 385M

3.9 slim (glibc) vertexai - - 4.62s 272M

Imports

vertexai.init
wrong
```
from google.cloud import aiplatform
aiplatform.init(project='your-project-id', location='us-central1')
```
correct
```
import vertexai

vertexai.init(
    project='your-project-id',
    location='us-central1'
)
```
install is google-cloud-aiplatform but import is vertexai — not google.cloud.aiplatform. Both packages coexist but vertexai is the high-level SDK. google.cloud.aiplatform is the low-level GAPIC client.

vertexai.generative_models (deprecated)

wrong

import vertexai
from vertexai.generative_models import GenerativeModel

vertexai.init(project='your-project', location='us-central1')
model = GenerativeModel('gemini-2.5-flash')
response = model.generate_content('Explain neural networks')
print(response.text)

correct

# Use google-genai for generative AI on Vertex AI
from google import genai

client = genai.Client(
    vertexai=True,
    project='your-project-id',
    location='us-central1'
)
response = client.models.generate_content(
    model='gemini-2.5-flash',
    contents='Explain neural networks'
)
print(response.text)

vertexai.generative_models deprecated June 24, 2025. Removed June 24, 2026. Migrate to google-genai with vertexai=True.

Application Default Credentials

wrong

import vertexai
vertexai.init(project='your-project-id', location='us-central1', credentials='your-api-key')

correct

import vertexai

# Uses ADC automatically — run: gcloud auth application-default login
vertexai.init(project='your-project-id', location='us-central1')

Vertex AI does not use API keys. Uses Google Cloud Application Default Credentials (ADC) or service account JSON. No api_key parameter exists on vertexai.init().

Quickstart reviewed last tested: 2026-04-23

Vertex AI Gemini call using google-genai with vertexai=True (recommended path from 2025 onwards).

# For Vertex AI generative AI — use google-genai
# pip install google-genai
from google import genai

client = genai.Client(
    vertexai=True,
    project='your-gcp-project-id',
    location='us-central1'
)

response = client.models.generate_content(
    model='gemini-2.5-flash',
    contents='What is Vertex AI?'
)
print(response.text)