Providers

Helicone

AISDK provides first-class support for Helicone 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 Helicone provider feature:

cargo add aisdk --features helicone

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

Create a Provider Instance

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

use aisdk::providers::Helicone;

let helicone = Helicone::chatgpt_4o_latest();

This initializes the provider with:

Basic Text Generation

Example using LanguageModelRequest for text generation.

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

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

    let helicone = Helicone::chatgpt_4o_latest();

    let response = LanguageModelRequest::builder()
        .model(helicone)
        .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 Helicone::builder()

API Key

let helicone = Helicone::<Chatgpt4oLatest>::builder()
    .api_key("your-api-key")
    .build()?;

If not specified, AISDK uses the HELICONE_API_KEY environment variable.

Base URL

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

let helicone = Helicone::<Chatgpt4oLatest>::builder()
    .base_url("https://ai-gateway.helicone.ai/v1")
    .build()?;

Path (Full URL Override)

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

let helicone = Helicone::<Chatgpt4oLatest>::builder()
    .path("https://full-url.example/v1/chat/completions")
    .build()?;

Provider Name

For logging, analytics, and observability.

let helicone = Helicone::<Chatgpt4oLatest>::builder()
    .provider_name("Helicone")
    .build()?;

Full Custom Configuration Example

let helicone = Helicone::<Chatgpt4oLatest>::builder()
    .api_key("your-api-key")
    .base_url("https://ai-gateway.helicone.ai/v1")
    .path("https://full-url.example/v1/chat/completions")
    .provider_name("Helicone")
    .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::Helicone;

// Specify model as a string at runtime
let helicone = Helicone::model_name("chatgpt-4o-latest");

Using Builder Pattern with Custom Settings

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

let helicone = Helicone::<DynamicModel>::builder()
	.model_name("chatgpt-4o-latest")
	.api_key("your-api-key")
	.base_url("https://ai-gateway.helicone.ai/v1")
	.path("https://full-url.example/v1/chat/completions")
	.provider_name("Helicone")
	.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 Helicone::chatgpt_4o_latest().

Next Steps