RLLM

RLLM is a Rust library that lets you use multiple LLM backends in a single project: OpenAI, Anthropic (Claude), Ollama, DeepSeek, xAI and Phind. With a unified API and builder style - similar to the Stripe experience - you can easily create chat or text completion requests without multiplying structures and crates.

Key Features

• Multi-backend: Manage OpenAI, Anthropic, Ollama, DeepSeek, xAI through a single entry point.
• Multi-step chains: Create multi-step chains with different backends at each step.
• Templates: Use templates to create complex prompts with variables.
• Builder pattern: Configure your LLM (model, temperature, max_tokens, timeouts...) with a few simple calls.
• Chat & Completions: Two unified traits (ChatProvider and CompletionProvider) to cover most use cases.
• Extensible: Easily add new backends.
• Rust-friendly: Designed with clear traits, unified error handling, and conditional compilation via features.
• Validation: Add validation to your requests to ensure the output is what you expect.
• Evaluation: Add evaluation to your requests to score the output of LLMs.

Installation

Simply add RLLM to your Cargo.toml:

[dependencies]
rllm = { version = "1.0.0", features = ["openai", "anthropic", "ollama"] }

Examples

Usage

Here's a basic example using OpenAI for chat completion. See the examples directory for other backends (Anthropic, Ollama, DeepSeek, xAI), embedding capabilities, and more advanced use cases.

use rllm::{
    builder::{LLMBackend, LLMBuilder},
    chat::{ChatMessage, ChatRole},
};

fn main() {
    let llm = LLMBuilder::new()
        .backend(LLMBackend::OpenAI) // or LLMBackend::Anthropic, LLMBackend::Ollama, LLMBackend::DeepSeek, LLMBackend::XAI, LLMBackend::Phind ...
        .api_key(std::env::var("OPENAI_API_KEY").unwrap_or("sk-TESTKEY".into()))
        .model("gpt-4o") // or model("claude-3-5-sonnet-20240620") or model("grok-2-latest") or model("deepseek-chat") or model("llama3.1") or model("Phind-70B") ...
        .max_tokens(1000)
        .temperature(0.7)
        .system("You are a helpful assistant.")
        .stream(false)
        .build()
        .expect("Failed to build LLM");
}

    let messages = vec![
        ChatMessage {
            role: ChatRole::User,
            content: "Tell me that you love cats".into(),
        },
        ChatMessage {
            role: ChatRole::Assistant,
            content:
                "I am an assistant, I cannot love cats but I can love dogs"
                    .into(),
        },
        ChatMessage {
            role: ChatRole::User,
            content: "Tell me that you love dogs in 2000 chars".into(),
        },
    ];

    let chat_resp = llm.chat(&messages);
    match chat_resp {
        Ok(text) => println!("Chat response:\n{}", text),
        Err(e) => eprintln!("Chat error: {}", e),
    }