coreweave_inference_gateway (Resource)

​

Example usage

resource "coreweave_inference_gateway" "example" {
  name  = "my-gateway"
  zones = ["US-EAST-04A"]

  auth = {
    coreweave = {}
  }

  routing = {
    body_based = {
      api_type = "API_TYPE_OPENAI"
    }
  }
}

​

Schema

​

Required

• auth (Attributes) The authentication configuration for the gateway. Exactly one of coreweave or weights_and_biases must be specified. (see below for nested schema)
• name (String) The human-readable name of the gateway. Capped at 38 characters so the derived public FQDN fits within the 64-character X.509 Common Name limit.
• routing (Attributes) The routing configuration for the gateway. Exactly one of body_based, header_based, or path_based must be specified. (see below for nested schema)
• zones (Set of String) The zones to make the gateway available in. Limits where deployments associated with the gateway may exist.

​

Optional

• endpoint_configuration (Attributes) Additional endpoint configuration options. (see below for nested schema)

​

Read-Only

• conditions (Attributes List) Detailed status conditions for the gateway. (see below for nested schema)
• created_at (String) RFC3339 timestamp of when the gateway was created.
• endpoints (Set of String) The endpoint URIs for the gateway.
• id (String) The unique identifier of the gateway.
• organization_id (String) The organization ID that owns the gateway.
• status (String) The current status of the gateway. See the Inference API overview for status values.
• updated_at (String) RFC3339 timestamp of when the gateway was last updated.

​

Nested Schema for auth

• coreweave (Attributes) Use CoreWeave IAM authentication. (see below for nested schema)
• weights_and_biases (Attributes) Use Weights & Biases authentication. (see below for nested schema)

​

Nested Schema for auth.coreweave

​

Nested Schema for auth.weights_and_biases

• api_key (String, Sensitive) The organization API key for Weights & Biases. Requires server_url to also be set.
• enable_rate_limiting (Boolean) Whether to enable Weights & Biases controlled rate limiting.
• enable_usage_reports (Boolean) Whether to send usage data to Weights & Biases.
• server_url (String) The Weights & Biases server URL. Defaults to the shared SaaS instance if not set.

​

Nested Schema for routing

• body_based (Attributes) Body-based routing configuration. Routes requests based on the request body content. (see below for nested schema)
• header_based (Attributes) Header-based routing configuration. Routes requests using a header value to match the model by name. (see below for nested schema)
• path_based (Attributes) Path-based routing configuration. Routes requests based on the URL path. (see below for nested schema)

​

Nested Schema for routing.body_based

• api_type (String) The well-known API type for routing. Must be one of: API_TYPE_OPENAI.

​

Nested Schema for routing.header_based

• header_name (String) The name of the header to use for routing.

​

Nested Schema for routing.path_based

​

Nested Schema for endpoint_configuration

• additional_dns (Set of String) Additional DNS names for the gateway endpoint. These DNS names must be manually configured to point to the gateway endpoint.
• allowed_source_ip_ranges (Set of String) Client source IP ranges permitted to reach the gateway, in CIDR notation (IPv4 or IPv6). Up to 50 entries; an empty list applies no source restriction. Use a full-length prefix for a single address (e.g. 203.0.113.5/32).
• edge_proxy_mode (String) How client traffic reaches the gateway. Must be one of: EDGE_PROXY_MODE_MANAGED (traffic is proxied through CoreWeave’s managed edge network: DDoS mitigation and WAF, with TLS terminated at the edge and the origin address kept out of public DNS) or EDGE_PROXY_MODE_DIRECT (clients connect straight to the gateway’s load balancer, with no edge filtering and the load balancer address published in DNS). When unset, the platform default (currently EDGE_PROXY_MODE_MANAGED) applies.

​

Nested Schema for conditions

• last_update_time (String) RFC3339 timestamp of the last condition transition.
• message (String) A human-readable message about the condition’s last transition.
• reason (String) A short, machine-readable reason for the condition’s last transition.
• status (String) The condition status (True, False, or Unknown).
• type (String) The condition type (e.g. Ready, Progressing).

​

Import

terraform import coreweave_inference_gateway.example {{gateway-id}}