Smallest AI LLM Adapter for RTP Agent

Smallest AI focuses on compact, highly efficient LLMs that can run with minimal latency and reduced computational requirements.

Installation

Add the adapter to your Go project:

go get github.com/cavos-io/rtp-agent/adapter/smallestai

Authentication

Set the required environment variables in your .env file. Refer to the Smallest AI developer documentation for acquiring the necessary API keys and tokens.

SMALLESTAI_API_KEY=your_api_key_here

Usage

Below is a basic conceptual example demonstrating how to initialize the Smallest AI LLM adapter within an RTP Agent session:

package main

import (
	"context"
	"log"
	"os"

	"github.com/cavos-io/rtp-agent/adapter/smallestai"
	"github.com/cavos-io/rtp-agent/core/agent"
)

func main() {
	ctx := context.Background()

	// Initialize the Smallest AI LLM adapter
	llmProvider, err := smallestai.NewProvider(
		os.Getenv("SMALLESTAI_API_KEY"),
	)
	if err != nil {
		log.Fatalf("failed to initialize smallestai adapter: %v", err)
	}

	// Create and configure the RTP agent session
	session := agent.NewSession(
		agent.WithLLM(llmProvider),
	)

	// Start the session
	if err := session.Start(ctx); err != nil {
		log.Fatalf("session failed: %v", err)
	}
}