Google Vertex AI provider implementation.

Supports Vertex AI's unified API for accessing multiple AI models including:

• Anthropic Claude models (claude-haiku-4-5, claude-sonnet-4-5, claude-opus-4-1)
• Google Gemini models (gemini-2.0-flash, gemini-2.5-flash, gemini-2.5-pro)
• Mistral AI models (mistral-medium-3, mistral-small-2503, codestral-2)
• Third-party MaaS models via OpenAI-compatible format:
  • GLM models (zai-org/glm-4.7-maas)
  • OpenAI OSS models (openai/gpt-oss-120b-maas, openai/gpt-oss-20b-maas)
• And more as Google adds them

Authentication

Vertex AI uses Google Cloud OAuth2 authentication with service accounts.

Service Account (Recommended)

# Option 1: Environment variables
export GOOGLE_APPLICATION_CREDENTIALS="/path/to/service-account.json"
export GOOGLE_CLOUD_PROJECT="your-project-id"
export GOOGLE_CLOUD_REGION="us-central1"

# Option 2: File path in options
ReqLLM.generate_text(
  "google-vertex:claude-haiku-4-5@20251001",
  "Hello",
  provider_options: [
    service_account_json: "/path/to/service-account.json",
    project_id: "your-project-id",
    region: "us-central1"
  ]
)

# Option 3: JSON string directly (no file needed)
ReqLLM.generate_text(
  "google-vertex:claude-haiku-4-5@20251001",
  "Hello",
  provider_options: [
    service_account_json: ~s({"client_email": "...", "private_key": "..."}),
    project_id: "your-project-id"
  ]
)

# Option 4: Pre-parsed map
ReqLLM.generate_text(
  "google-vertex:claude-haiku-4-5@20251001",
  "Hello",
  provider_options: [
    service_account_json: %{"client_email" => "...", "private_key" => "..."},
    project_id: "your-project-id"
  ]
)

Access Token

ReqLLM.generate_text(
  "google-vertex:gemini-2.5-flash",
  "Hello from Vertex",
  provider_options: [
    access_token: "your-access-token",
    project_id: "your-project-id",
    region: "us-central1"
  ]
)

Examples

# Simple text generation with Claude on Vertex
{:ok, response} = ReqLLM.generate_text(
  "google-vertex:claude-haiku-4-5@20251001",
  "Hello!"
)

# Streaming
{:ok, response} = ReqLLM.stream_text(
  "google-vertex:claude-haiku-4-5@20251001",
  "Tell me a story"
)

Extending for New Models

To add support for a new model family:

1. Add the model family to @model_families
2. Implement the formatter module (e.g., ReqLLM.Providers.GoogleVertex.Gemini)
3. The formatter needs:
   • format_request/3 - Convert ReqLLM context to provider format
   • parse_response/2 - Convert provider response to ReqLLM format
   • extract_usage/2 - Extract usage information