Vertex AI¶
Google Cloud Vertex AI platform for accessing Gemini and other models.
Configuration¶
PROVIDER=vertexai
PROJECT=your-gcp-project-id
LOCATION=us-central1
MODEL=gemini-1.5-pro
TEMPERATURE=0.3
Key Parameters¶
| Parameter | Description | Default |
|---|---|---|
PROJECT |
GCP project ID (required) | - |
LOCATION |
GCP region (required) | - |
MODEL |
Model to use | - |
TEMPERATURE |
Randomness (0.0-1.0) | 0.3 |
TOP_P |
Nucleus sampling | 1.0 |
TOP_K |
Top-k sampling | - |
MAX_TOKENS |
Max response tokens | 1024 |
TIMEOUT |
Request timeout (seconds) | 60 |
Features¶
- ✅ Streaming responses
- ✅ Function calling
- ✅ MCP support
- ✅ Vision capabilities
- ✅ Enterprise features
- ✅ Multi-modal support
Authentication¶
Vertex AI requires Google Cloud authentication. Set up authentication using one of these methods:
# Application Default Credentials
gcloud auth application-default login
# Service Account Key
export GOOGLE_APPLICATION_CREDENTIALS=/path/to/service-account.json
# Workload Identity (for GKE)
# Configured automatically in GKE environment
Important Notes¶
- Based on Gemini provider implementation
- Requires GCP project and location configuration
- Automatic authentication via Google Cloud SDK
- Enterprise-grade security and compliance
- Regional model availability may vary
- Billing through Google Cloud Console