Providers

altimate supports 35+ LLM providers. Configure them in the provider section of your config file.

Provider Configuration

Each provider has a key in the provider object:

{
  "provider": {
    "<provider-name>": {
      "apiKey": "{env:API_KEY}",
      "baseURL": "https://custom.endpoint.com/v1",
      "headers": {
        "X-Custom-Header": "value"
      }
    }
  }
}

Tip

Use {env:...} substitution for API keys so you never commit secrets to version control.

Anthropic

{
  "provider": {
    "anthropic": {
      "apiKey": "{env:ANTHROPIC_API_KEY}"
    }
  },
  "model": "anthropic/claude-sonnet-4-6"
}

Available models: claude-opus-4-6, claude-sonnet-4-6, claude-haiku-4-5-20251001

OpenAI

{
  "provider": {
    "openai": {
      "apiKey": "{env:OPENAI_API_KEY}"
    }
  },
  "model": "openai/gpt-4o"
}

AWS Bedrock

{
  "provider": {
    "bedrock": {
      "region": "us-east-1",
      "accessKeyId": "{env:AWS_ACCESS_KEY_ID}",
      "secretAccessKey": "{env:AWS_SECRET_ACCESS_KEY}"
    }
  },
  "model": "bedrock/anthropic.claude-sonnet-4-6-v1"
}

Uses the standard AWS credential chain. Set AWS_PROFILE or provide credentials directly.

Note

If you have AWS SSO or IAM roles configured, Bedrock will use your default credential chain automatically, so no explicit keys are needed.

Azure OpenAI

{
  "provider": {
    "azure": {
      "apiKey": "{env:AZURE_OPENAI_API_KEY}",
      "baseURL": "https://your-resource.openai.azure.com/openai/deployments/your-deployment"
    }
  },
  "model": "azure/gpt-4o"
}

Google (Gemini)

{
  "provider": {
    "google": {
      "apiKey": "{env:GOOGLE_API_KEY}"
    }
  },
  "model": "google/gemini-2.5-pro"
}

Google Vertex AI

{
  "provider": {
    "google-vertex": {
      "project": "my-gcp-project",
      "location": "us-central1"
    }
  },
  "model": "google-vertex/gemini-2.5-pro"
}

Uses Google Cloud Application Default Credentials. Authenticate with:

gcloud auth application-default login

The project and location fields can also be set via environment variables:

Field	Environment Variables (checked in order)
project	GOOGLE_CLOUD_PROJECT, GCP_PROJECT, GCLOUD_PROJECT
location	GOOGLE_VERTEX_LOCATION, GOOGLE_CLOUD_LOCATION, VERTEX_LOCATION

If location is not set, it defaults to us-central1.

Tip

You can also access Anthropic models through Vertex AI using the google-vertex provider (e.g., google-vertex/claude-sonnet-4-6).

Ollama (Local)

{
  "provider": {
    "ollama": {
      "baseURL": "http://localhost:11434"
    }
  },
  "model": "ollama/llama3.1"
}

No API key needed. Runs entirely on your local machine.

Info

Make sure Ollama is running before starting altimate. Install it from ollama.com and pull your desired model with ollama pull llama3.1.

OpenRouter

{
  "provider": {
    "openrouter": {
      "apiKey": "{env:OPENROUTER_API_KEY}"
    }
  },
  "model": "openrouter/anthropic/claude-sonnet-4-6"
}

Access 150+ models through a single API key.

Copilot

{
  "provider": {
    "copilot": {}
  },
  "model": "copilot/gpt-4o"
}

Uses your GitHub Copilot subscription. Authenticate with altimate auth.

Custom / OpenAI-Compatible

Any OpenAI-compatible endpoint can be used as a provider:

{
  "provider": {
    "my-provider": {
      "api": "openai",
      "baseURL": "https://my-llm-proxy.example.com/v1",
      "apiKey": "{env:MY_API_KEY}"
    }
  },
  "model": "my-provider/my-model"
}

Tip

This works with any service that exposes an OpenAI-compatible chat completions API, including vLLM, LiteLLM, and self-hosted inference servers.

Model Selection

Set your default model and a smaller model for lightweight tasks:

{
  "model": "anthropic/claude-sonnet-4-6",
  "small_model": "anthropic/claude-haiku-4-5-20251001"
}

The small_model is used for lightweight tasks like summarization and context compaction.

Provider Options Reference

Field	Type	Description
apiKey	string	API key (supports {env:...} and {file:...})
baseURL	string	Custom API endpoint URL
api	string	API type (e.g., "openai" for compatible endpoints)
headers	object	Custom HTTP headers to include with requests
region	string	AWS region (Bedrock only)
accessKeyId	string	AWS access key (Bedrock only)
secretAccessKey	string	AWS secret key (Bedrock only)
project	string	GCP project ID (Google Vertex AI only)
location	string	GCP region (Google Vertex AI only, default: us-central1)