ExuluDefaultAgents

Overview

ExuluDefaultAgents provides ready-to-use ExuluProvider configurations for popular AI providers supported by the Vercel AI SDK. These default agents are convenience wrappers that make it easy to get started with different providers without having to configure each one manually.

ExuluDefaultAgents are merely implementations of AI SDK providers. You can easily create agents for any provider supported by the AI SDK by using the ExuluProvider class directly. See the AI SDK providers documentation for the full list of available providers.

Available default agents

ExuluDefaultAgents includes pre-configured agents for the most popular providers:

OpenAI

GPT-4, GPT-4 Turbo, GPT-3.5, and more

Anthropic

Claude Opus, Sonnet, and Haiku models

Google

Gemini Pro, Gemini Flash models

Custom providers

Easy to add any AI SDK provider

Quick start

import { ExuluDefaultAgents } from "@exulu/backend";

// Use a pre-configured agent (example - actual API depends on implementation)
const openaiAgent = ExuluDefaultAgents.openai;
const anthropicAgent = ExuluDefaultAgents.anthropic;
const googleAgent = ExuluDefaultAgents.google;

Default agents still require authentication credentials. Use ExuluVariables to manage API keys securely.

Creating custom agents with AI SDK providers

While ExuluDefaultAgents provides convenient defaults, you can create agents for any provider supported by the AI SDK:

OpenAI

import { ExuluProvider } from "@exulu/backend";
import { createOpenAI } from "@ai-sdk/openai";

const openaiAgent = new ExuluProvider({
  id: "openai_gpt4",
  name: "OpenAI GPT-4",
  type: "agent",
  description: "GPT-4 powered agent",
  provider: "openai",
  config: {
    name: "gpt-4o",
    model: {
      create: ({ apiKey }) => createOpenAI({ apiKey })("gpt-4o")
    },
    instructions: "You are a helpful AI assistant."
  },
  capabilities: {
    text: true,
    images: [".png", ".jpg", ".jpeg", ".gif", ".webp"],
    files: [".pdf", ".txt"],
    audio: [],
    video: []
  }
});

Available models: gpt-4o, gpt-4o-mini, gpt-4-turbo, gpt-4, gpt-3.5-turbo OpenAI AI SDK Documentation

Anthropic

import { ExuluProvider } from "@exulu/backend";
import { createAnthropic } from "@ai-sdk/anthropic";

const anthropicAgent = new ExuluProvider({
  id: "anthropic_claude",
  name: "Claude Opus 4",
  type: "agent",
  description: "Claude Opus 4 powered agent",
  provider: "anthropic",
  config: {
    name: "claude-opus-4-20250514",
    model: {
      create: ({ apiKey }) => createAnthropic({ apiKey })("claude-opus-4-20250514")
    },
    instructions: "You are Claude, a helpful AI assistant."
  },
  capabilities: {
    text: true,
    images: [".png", ".jpg", ".jpeg", ".gif", ".webp"],
    files: [".pdf", ".txt"],
    audio: [],
    video: []
  }
});

Available models: claude-opus-4-20250514, claude-sonnet-4-20250514, claude-3-5-sonnet-20241022, claude-3-opus-20240229, claude-3-haiku-20240307 Anthropic AI SDK Documentation

Google Generative AI

import { ExuluProvider } from "@exulu/backend";
import { createGoogleGenerativeAI } from "@ai-sdk/google";

const googleAgent = new ExuluProvider({
  id: "google_gemini",
  name: "Google Gemini Pro",
  type: "agent",
  description: "Gemini Pro powered agent",
  provider: "google",
  config: {
    name: "gemini-2.0-flash-exp",
    model: {
      create: ({ apiKey }) => createGoogleGenerativeAI({ apiKey })("gemini-2.0-flash-exp")
    },
    instructions: "You are a helpful AI assistant."
  },
  capabilities: {
    text: true,
    images: [".png", ".jpg", ".jpeg", ".gif", ".webp"],
    files: [],
    audio: [],
    video: []
  }
});

Available models: gemini-2.0-flash-exp, gemini-1.5-pro, gemini-1.5-flash, gemini-1.0-pro Google AI SDK Documentation

Mistral

import { ExuluProvider } from "@exulu/backend";
import { createMistral } from "@ai-sdk/mistral";

const mistralAgent = new ExuluProvider({
  id: "mistral_large",
  name: "Mistral Large",
  type: "agent",
  description: "Mistral Large powered agent",
  provider: "mistral",
  config: {
    name: "mistral-large-latest",
    model: {
      create: ({ apiKey }) => createMistral({ apiKey })("mistral-large-latest")
    },
    instructions: "You are a helpful AI assistant."
  },
  capabilities: {
    text: true,
    images: [],
    files: [],
    audio: [],
    video: []
  }
});

Available models: mistral-large-latest, mistral-small-latest, codestral-latest Mistral AI SDK Documentation

Cohere

import { ExuluProvider } from "@exulu/backend";
import { createCohere } from "@ai-sdk/cohere";

const cohereAgent = new ExuluProvider({
  id: "cohere_command",
  name: "Cohere Command R+",
  type: "agent",
  description: "Cohere Command R+ powered agent",
  provider: "cohere",
  config: {
    name: "command-r-plus",
    model: {
      create: ({ apiKey }) => createCohere({ apiKey })("command-r-plus")
    },
    instructions: "You are a helpful AI assistant."
  },
  capabilities: {
    text: true,
    images: [],
    files: [],
    audio: [],
    video: []
  }
});

Available models: command-r-plus, command-r, command-light Cohere AI SDK Documentation

Groq

import { ExuluAgent } from "@exulu/backend";
import { createGroq } from "@ai-sdk/groq";

const groqAgent = new ExuluAgent({
  id: "groq_llama",
  name: "Groq Llama 3",
  type: "agent",
  description: "Groq-powered Llama 3 agent",
  provider: "groq",
  config: {
    name: "llama-3.1-70b-versatile",
    model: {
      create: ({ apiKey }) => createGroq({ apiKey })("llama-3.1-70b-versatile")
    },
    instructions: "You are a helpful AI assistant."
  },
  capabilities: {
    text: true,
    images: [],
    files: [],
    audio: [],
    video: []
  }
});

Available models: llama-3.1-70b-versatile, llama-3.1-8b-instant, mixtral-8x7b-32768 Groq AI SDK Documentation

Azure OpenAI

import { ExuluAgent } from "@exulu/backend";
import { createAzure } from "@ai-sdk/azure";

const azureAgent = new ExuluAgent({
  id: "azure_gpt4",
  name: "Azure GPT-4",
  type: "agent",
  description: "Azure OpenAI GPT-4 agent",
  provider: "azure",
  config: {
    name: "gpt-4",
    model: {
      create: ({ apiKey }) => createAzure({
        apiKey,
        resourceName: "your-resource-name",
      })("gpt-4-deployment-name")
    },
    instructions: "You are a helpful AI assistant."
  },
  capabilities: {
    text: true,
    images: [".png", ".jpg", ".jpeg", ".gif", ".webp"],
    files: [".pdf", ".txt"],
    audio: [],
    video: []
  }
});

Azure OpenAI AI SDK Documentation

Using with ExuluVariables

Combine custom agents with ExuluVariables for secure credential management:

import { ExuluAgent, ExuluVariables } from "@exulu/backend";
import { createOpenAI } from "@ai-sdk/openai";
import { createAnthropic } from "@ai-sdk/anthropic";

// Retrieve API keys from variables
const openaiKey = await ExuluVariables.get("openai_api_key");
const anthropicKey = await ExuluVariables.get("anthropic_api_key");

// Create agents with retrieved keys
const openaiAgent = new ExuluAgent({
  id: "openai_agent",
  name: "OpenAI Assistant",
  type: "agent",
  description: "OpenAI powered assistant",
  provider: "openai",
  authenticationInformation: openaiKey,
  config: {
    name: "gpt-4o",
    model: {
      create: ({ apiKey }) => createOpenAI({ apiKey: apiKey || openaiKey })("gpt-4o")
    },
    instructions: "You are a helpful assistant."
  },
  capabilities: { text: true, images: [], files: [], audio: [], video: [] }
});

const anthropicAgent = new ExuluAgent({
  id: "anthropic_agent",
  name: "Claude Assistant",
  type: "agent",
  description: "Claude powered assistant",
  provider: "anthropic",
  authenticationInformation: anthropicKey,
  config: {
    name: "claude-opus-4-20250514",
    model: {
      create: ({ apiKey }) => createAnthropic({ apiKey: apiKey || anthropicKey })("claude-opus-4-20250514")
    },
    instructions: "You are Claude, a helpful assistant."
  },
  capabilities: { text: true, images: [], files: [], audio: [], video: [] }
});

Multi-provider setup

Create a registry of agents for different providers:

import { ExuluAgent, ExuluVariables } from "@exulu/backend";
import { createOpenAI } from "@ai-sdk/openai";
import { createAnthropic } from "@ai-sdk/anthropic";
import { createGoogleGenerativeAI } from "@ai-sdk/google";

async function createAgentRegistry() {
  // Retrieve all API keys
  const [openaiKey, anthropicKey, googleKey] = await Promise.all([
    ExuluVariables.get("openai_api_key"),
    ExuluVariables.get("anthropic_api_key"),
    ExuluVariables.get("google_api_key")
  ]);

  return {
    openai: new ExuluAgent({
      id: "openai_gpt4",
      name: "OpenAI GPT-4",
      type: "agent",
      provider: "openai",
      authenticationInformation: openaiKey,
      config: {
        name: "gpt-4o",
        model: {
          create: ({ apiKey }) => createOpenAI({ apiKey: apiKey || openaiKey })("gpt-4o")
        },
        instructions: "You are a helpful assistant."
      },
      capabilities: { text: true, images: [], files: [], audio: [], video: [] }
    }),

    anthropic: new ExuluAgent({
      id: "anthropic_opus",
      name: "Claude Opus 4",
      type: "agent",
      provider: "anthropic",
      authenticationInformation: anthropicKey,
      config: {
        name: "claude-opus-4-20250514",
        model: {
          create: ({ apiKey }) => createAnthropic({ apiKey: apiKey || anthropicKey })("claude-opus-4-20250514")
        },
        instructions: "You are Claude, a helpful assistant."
      },
      capabilities: { text: true, images: [], files: [], audio: [], video: [] }
    }),

    google: new ExuluAgent({
      id: "google_gemini",
      name: "Google Gemini",
      type: "agent",
      provider: "google",
      authenticationInformation: googleKey,
      config: {
        name: "gemini-2.0-flash-exp",
        model: {
          create: ({ apiKey }) => createGoogleGenerativeAI({ apiKey: apiKey || googleKey })("gemini-2.0-flash-exp")
        },
        instructions: "You are a helpful assistant."
      },
      capabilities: { text: true, images: [], files: [], audio: [], video: [] }
    })
  };
}

// Use the registry
const agents = await createAgentRegistry();

const response = await agents.openai.generateSync({
  prompt: "Hello!",
  agentInstance: await loadAgent("openai_gpt4"),
  statistics: { label: "openai", trigger: "api" }
});

Dynamic provider selection

Allow users to choose their preferred provider:

import { ExuluAgent, ExuluVariables } from "@exulu/backend";
import { createOpenAI } from "@ai-sdk/openai";
import { createAnthropic } from "@ai-sdk/anthropic";

async function createAgentForProvider(provider: "openai" | "anthropic") {
  const providerConfigs = {
    openai: {
      variableName: "openai_api_key",
      modelName: "gpt-4o",
      createFn: createOpenAI
    },
    anthropic: {
      variableName: "anthropic_api_key",
      modelName: "claude-opus-4-20250514",
      createFn: createAnthropic
    }
  };

  const config = providerConfigs[provider];
  const apiKey = await ExuluVariables.get(config.variableName);

  return new ExuluAgent({
    id: `${provider}_agent`,
    name: `${provider} Agent`,
    type: "agent",
    provider,
    authenticationInformation: apiKey,
    config: {
      name: config.modelName,
      model: {
        create: ({ apiKey: key }) => config.createFn({ apiKey: key || apiKey })(config.modelName)
      },
      instructions: "You are a helpful assistant."
    },
    capabilities: { text: true, images: [], files: [], audio: [], video: [] }
  });
}

// User selects provider
const selectedProvider = "openai"; // or "anthropic"
const agent = await createAgentForProvider(selectedProvider);

Explore more providers

The AI SDK supports many more providers beyond the examples above:

OpenAI

GPT-4, GPT-3.5

Anthropic

Claude models

Google

Gemini models

Mistral

Mistral Large, Small

Cohere

Command R+

Groq

Fast LLaMA inference

Azure OpenAI

Enterprise OpenAI

Vertex AI

Google Cloud AI

Amazon Bedrock

AWS AI models

Visit the AI SDK Providers documentation for the complete list of supported providers and their configurations.

Community providers

The AI SDK also supports community-maintained providers:

Ollama: Run local models
Replicate: Access thousands of models
Fireworks AI: Fast inference
Together AI: Open-source models
Perplexity: Search-augmented models
xAI (Grok): xAI’s Grok models
And many more…

View all community providers

Best practices

Start with defaults: Use ExuluDefaultAgents to get started quickly, then customize as needed.

Secure credentials: Always use ExuluVariables to manage API keys securely.

Provider selection: Choose providers based on your needs: cost, latency, model capabilities, and data privacy requirements.

Rate limits: Different providers have different rate limits. Configure ExuluQueues accordingly.

Next steps

ExuluAgent

Learn about ExuluAgent configuration

ExuluVariables

Manage API keys securely

AI SDK Providers

Explore all available providers

AI SDK Docs

Full AI SDK documentation

Getting started

Core classes

Frontend

GraphQL API

Core Types

​Overview

​Available default agents

OpenAI

Anthropic

Google

Custom providers

​Quick start

​Creating custom agents with AI SDK providers

​OpenAI

​Anthropic

​Google Generative AI

​Mistral

​Cohere

​Groq

​Azure OpenAI

​Using with ExuluVariables

​Multi-provider setup

​Dynamic provider selection

​Explore more providers

OpenAI

Anthropic

Google

Mistral

Cohere

Groq

Azure OpenAI

Vertex AI

Amazon Bedrock

​Community providers

​Best practices

​Next steps

ExuluAgent

ExuluVariables

AI SDK Providers

AI SDK Docs

Overview

Available default agents

Quick start

Creating custom agents with AI SDK providers

OpenAI

Anthropic

Google Generative AI

Mistral

Cohere

Groq

Azure OpenAI

Using with ExuluVariables

Multi-provider setup

Dynamic provider selection

Explore more providers

Community providers

Best practices

Next steps