> For the complete documentation index, see llms.txt.
Skip to main content

Check out Port for yourself ➜ 

Provider guides overview

This section provides step-by-step setup instructions for each LLM provider supported by Port. Each guide walks you through provider-specific configuration, authentication, and registration using Port's API.

Supported providers​

Port supports the following LLM providers. Select a provider below to view its specific setup guide:

Direct provider connections​

  • OpenAI - Connect directly to OpenAI's API with your API key. Supports gpt-5 and other OpenAI models.
  • Anthropic - Connect directly to Anthropic's API with your API key. Supports Claude models including claude-sonnet-4-20250514, claude-haiku-4-5-20251001, claude-sonnet-4-5-20250929, claude-opus-4-5-20251101, and claude-opus-4-6.

Cloud provider integrations​

  • Azure OpenAI - Connect to OpenAI models hosted on Azure with your Azure OpenAI deployment. Supports gpt-5 and other OpenAI models deployed in your Azure subscription.
  • Azure Anthropic - Connect to Anthropic models hosted on Azure. Supports Claude models including claude-sonnet-4-20250514, claude-haiku-4-5-20251001, claude-sonnet-4-5-20250929, claude-opus-4-5-20251101, and claude-opus-4-6.
  • AWS Bedrock - Connect to AI models through AWS Bedrock with IAM-based authentication. Supports Claude models through Bedrock's serverless inference.
  • Vertex Gemini - Connect to Google's Gemini models on Vertex AI. Supports both service account and Express Mode authentication.
  • Vertex Anthropic - Connect to Claude models hosted on Google Vertex AI through the Anthropic partner integration.

Proxy and gateway integrations​

  • OpenAI compatible - Connect to any HTTP endpoint that implements the OpenAI chat completions API. This includes LiteLLM proxies, custom gateways, and other OpenAI-compatible services.

General setup flow​

All provider guides follow this general flow:

  1. Configure provider policies and settings - Some providers (like AWS Bedrock) require additional configuration in the provider's console before you can use them with Port. This step is optional for providers that only need API keys.
  2. Store API keys in secrets - Store your provider credentials securely in Port's secrets system. See Step 2: Store API Keys in Secrets in the main setup guide.
  3. Register the provider - Use the Create or connect an LLM provider API to configure and validate your provider connection.
  4. Set as default - Optionally set your provider as the organization default using the UI (Builder → Organization Settings → AI tab) or the Change default LLM provider and model API.

Validation and troubleshooting​

For detailed information about connection validation, troubleshooting common issues, and understanding the configuration hierarchy, see Setup & Configuration.

Next steps​

Choose a provider from the list above to begin setup, or review the LLM Provider Management overview to understand how providers work in Port.