Vercel AI Gateway

AISDK provides first-class support for Vercel AI Gateway 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 Vercel AI Gateway provider feature:

cargo add aisdk --features vercel

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

Create a Provider Instance

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

use aisdk::providers::Vercel;

let vercel = Vercel::openai_gpt_4o_mini();

This initializes the provider with:

Model: "openai/gpt-4o-mini"
API key from environment (if set with AI_GATEWAY_API_KEY)
Vercel AI Gateway's default base URL (https://ai-gateway.vercel.sh/)

Basic Text Generation

Example using LanguageModelRequest for text generation.

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

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

    let vercel = Vercel::openai_gpt_4o_mini();

    let response = LanguageModelRequest::builder()
        .model(vercel)
        .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 Vercel::builder()

API Key

let vercel = Vercel::<OpenaiGpt4oMini>::builder()
    .api_key("your-api-key")
    .build()?;

If not specified, AISDK uses the AI_GATEWAY_API_KEY environment variable.

Base URL

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

let vercel = Vercel::<OpenaiGpt4oMini>::builder()
    .base_url("https://ai-gateway.vercel.sh/")
    .build()?;

Provider Name

For logging, analytics, and observability.

let vercel = Vercel::<OpenaiGpt4oMini>::builder()
    .provider_name("Vercel AI Gateway")
    .build()?;

Full Custom Configuration Example

let vercel = Vercel::<OpenaiGpt4oMini>::builder()
    .api_key("your-api-key")
    .base_url("https://ai-gateway.vercel.sh/")
    .provider_name("Vercel AI Gateway")
    .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::Vercel;

// Specify model as a string at runtime
let vercel = Vercel::model_name("openai/gpt-4o-mini");

Using Builder Pattern with Custom Settings

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

let vercel = Vercel::<DynamicModel>::builder()
    .model_name("openai/gpt-4o-mini")
    .api_key("your-api-key")
		.base_url("https://ai-gateway.vercel.sh/")
    .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 Vercel::openai_gpt_4o_mini().

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.

Vercel AI Gateway

On this page