OpenRouter

A provider block with kind = "openrouter" configures the OpenRouter provider, which proxies hundreds of models from many providers through a single unified API. This is useful when you want to switch between models from different providers without managing multiple API keys or changing your agent configuration.

Unlike the openai kind, OpenRouter does not have a fixed model catalogue. The models list accepts any model identifier that OpenRouter supports, in the form provider/model-name, such as openai/gpt-4o or anthropic/claude-3.5-sonnet.

Fields

When config.api_key is unset, the key is read from the OPENROUTER_API_KEY environment variable. The agent fails to start when neither is set.

ModelEntry

ProviderConfig

HeaderValue

A header value is one of:

config {
  headers = {
    "X-Team"      = "platform"
    "X-Tenant-Id" = expression("ctx.tenant_id")
  }
}

For the values an expression can read and how evaluation fails, see Send custom headers.

InferenceParams

The params block sets inference parameter defaults for all models under this provider. All fields are optional and support runtime().

Example

provider "openrouter" {
  kind = "openrouter"
  models = [
    "openai/gpt-4o-mini",
    {
      name = "anthropic/claude-3.5-sonnet"
      params {
        max_tokens  = 8192
        temperature = 0.7
      }
    }
  ]

  config {
    api_key = secret(runtime("OPENROUTER_API_KEY"))
  }

  params {
    max_tokens  = 2048
    temperature = 0.5
  }
}

agent "assistant" {
  graph {
    type = "react"
  }

  model {
    provider = "openrouter"
    name     = "openai/gpt-4o-mini"
  }
}