CloudflareAiGateway

AISDK provides first-class support for CloudflareAiGateway with fully typed model APIs. Model capabilities are enforced at compile time using Rust's type system. This prevents model capability mismatches and guarantees the selected model is valid for the task (e.g. tool calling).

Installation

Enable the CloudflareAiGateway provider feature:

cargo add aisdk --features cloudflare-ai-gateway

This installs AISDK with the CloudflareAiGateway provider enabled. Once you have enabled the CloudflareAiGateway provider, you can use all aisdk features with it.

Create a Provider Instance

To create a provider instance, call CloudflareAiGateway::model_name(), where model_name is the CloudflareAiGateway model you want to use. Model names are exposed as snake-case methods.

use aisdk::providers::CloudflareAiGateway;

let cloudflare-ai-gateway = CloudflareAiGateway::anthropic_claude_3_5_haiku();

This initializes the provider with:

Model: "anthropic/claude-3-5-haiku"
API key from environment (if set with CLOUDFLARE_API_TOKEN)
CloudflareAiGateway's default base URL (https://gateway.ai.cloudflare.com/v1/${CLOUDFLARE_ACCOUNT_ID}/${CLOUDFLARE_GATEWAY_ID}/compat/)

Basic Text Generation

Example using LanguageModelRequest for text generation.

use aisdk::{
    core::LanguageModelRequest,
    providers::CloudflareAiGateway,
};

#[tokio::main]
async fn main() -> Result<(), Box<dyn std::error::Error>> {

    let cloudflare-ai-gateway = CloudflareAiGateway::anthropic_claude_3_5_haiku();

    let response = LanguageModelRequest::builder()
        .model(cloudflare-ai-gateway)
        .prompt("Write a short poem about Rust.")
        .build()
        .generate_text()
        .await?;

    println!("Response text: {:?}", response.text());
    Ok(())
}

Provider Settings

You can customize provider configuration using CloudflareAiGateway::builder()

API Key

let cloudflare-ai-gateway = CloudflareAiGateway::<AnthropicClaude35Haiku>::builder()
    .api_key("your-api-key")
    .build()?;

If not specified, AISDK uses the CLOUDFLARE_API_TOKEN environment variable.

Base URL

Useful when routing through a proxy, gateway, or self-hosted compatible endpoint.

let cloudflare-ai-gateway = CloudflareAiGateway::<AnthropicClaude35Haiku>::builder()
    .base_url("https://gateway.ai.cloudflare.com/v1/${CLOUDFLARE_ACCOUNT_ID}/${CLOUDFLARE_GATEWAY_ID}/compat/")
    .build()?;

Path (Full URL Override)

Use .path(...) to override the full request URL instead of only the base URL.

let cloudflare-ai-gateway = CloudflareAiGateway::<AnthropicClaude35Haiku>::builder()
    .path("https://full-url.example/v1/chat/completions")
    .build()?;

Provider Name

For logging, analytics, and observability.

let cloudflare-ai-gateway = CloudflareAiGateway::<AnthropicClaude35Haiku>::builder()
    .provider_name("CloudflareAiGateway")
    .build()?;

Full Custom Configuration Example

let cloudflare-ai-gateway = CloudflareAiGateway::<AnthropicClaude35Haiku>::builder()
    .api_key("your-api-key")
    .base_url("https://gateway.ai.cloudflare.com/v1/${CLOUDFLARE_ACCOUNT_ID}/${CLOUDFLARE_GATEWAY_ID}/compat/")
    .path("https://full-url.example/v1/chat/completions")
    .provider_name("CloudflareAiGateway")
    .build()?;

Dynamic Model Selection

For runtime model selection (e.g., loading models from config files), use DynamicModel:

Using model_name() Method with Default Settings

use aisdk::providers::CloudflareAiGateway;

// Specify model as a string at runtime
let cloudflare-ai-gateway = CloudflareAiGateway::model_name("anthropic/claude-3-5-haiku");

Using Builder Pattern with Custom Settings

use aisdk::{
    core::DynamicModel,
    providers::CloudflareAiGateway,
};

let cloudflare-ai-gateway = CloudflareAiGateway::<DynamicModel>::builder()
	.model_name("anthropic/claude-3-5-haiku")
	.api_key("your-api-key")
	.base_url("https://gateway.ai.cloudflare.com/v1/${CLOUDFLARE_ACCOUNT_ID}/${CLOUDFLARE_GATEWAY_ID}/compat/")
	.path("https://full-url.example/v1/chat/completions")
	.provider_name("CloudflareAiGateway")
	.build()?;

Warning: When using DynamicModel, model capabilities are not validated at compile time. This means there's no guarantee the model supports requested features (e.g., tool calls, structured output). For compile-time safety, use the typed methods like CloudflareAiGateway::anthropic_claude_3_5_haiku().

Next Steps

Take a deeper look at text generation features Generating Text / Streaming Text
Explore Structured Output for reliable agent data.
Learn how to create Custom Tools.
Learn more about Agents.

CloudflareAiGateway

On this page