Rust

Visao Geral

O Floopy e um gateway sem SDK. O crate async-openai suporta URLs base personalizadas, entao voce pode rotear todas as requisicoes pelo Floopy sem dependencias extras. Voce ganha cache, rate limiting, fallbacks e observabilidade de graca.

Instalacao

Adicione ao seu Cargo.toml:

cargo add async-openai tokio

Configuracao

use async_openai::{Client, config::OpenAIConfig};

let config = OpenAIConfig::new()
    .with_api_base("https://api.floopy.ai/v1")
    .with_api_key("fp_your_api_key"); // starts with fp_

let client = Client::with_config(config);

Defina sua chave de API do Floopy (comeca com fp_). Voce pode criar uma no dashboard.

Requisicao Basica

use async_openai::types::{
    ChatCompletionRequestUserMessageArgs,
    CreateChatCompletionRequestArgs,
};

let request = CreateChatCompletionRequestArgs::default()
    .model("gpt-4o")
    .messages(vec![
        ChatCompletionRequestUserMessageArgs::default()
            .content("Explain quantum computing in one sentence.")
            .build()?
            .into(),
    ])
    .build()?;

let response = client.chat().create(request).await?;

println!("{}", response.choices[0].message.content.as_ref().unwrap());

Streaming

use futures::StreamExt;

let request = CreateChatCompletionRequestArgs::default()
    .model("gpt-4o")
    .messages(vec![
        ChatCompletionRequestUserMessageArgs::default()
            .content("Write a short poem.")
            .build()?
            .into(),
    ])
    .stream(true)
    .build()?;

let mut stream = client.chat().create_stream(request).await?;

while let Some(result) = stream.next().await {
    match result {
        Ok(response) => {
            for choice in &response.choices {
                if let Some(ref content) = choice.delta.content {
                    print!("{}", content);
                }
            }
        }
        Err(e) => eprintln!("Error: {}", e),
    }
}

Headers Personalizados

Passe headers especificos do Floopy configurando um cliente reqwest personalizado:

use reqwest::header::{HeaderMap, HeaderValue};

let mut headers = HeaderMap::new();
headers.insert("Floopy-Cache", HeaderValue::from_static("semantic"));
headers.insert("floopy-property-environment", HeaderValue::from_static("production"));
headers.insert("floopy-fallback", HeaderValue::from_static("claude-sonnet-4-20250514"));

let http_client = reqwest::Client::builder()
    .default_headers(headers)
    .build()?;

let config = OpenAIConfig::new()
    .with_api_base("https://api.floopy.ai/v1")
    .with_api_key("fp_your_api_key");

let client = Client::with_config(config).with_http_client(http_client);

Header	Descricao
`Floopy-Cache`	Estrategia de cache: `semantic` ou `exact`
`floopy-property-*`	Anexe metadados personalizados para filtragem no dashboard
`floopy-fallback`	Modelo de fallback se o provedor principal falhar
`floopy-session-id`	Agrupe requisicoes relacionadas em uma sessao
`floopy-user-id`	Associe requisicoes a um usuario final

Consulte a Referencia de Headers para a lista completa.