Python JavaScript

LLM Providers (JavaScript/TypeScript)¶

GraphBit supports multiple Large Language Model providers through a unified configuration interface. This guide covers configuration for each supported provider.

Supported Providers¶

GraphBit supports these LLM providers: - OpenAI - GPT models including GPT-4o - Azure OpenAI - Enterprise-grade OpenAI models - Anthropic - Claude models - MistralAI - Mistral models - ByteDance ModelArk - ByteDance Seed models - OpenRouter - Unified access to 400+ models - Perplexity - Search-enabled models - DeepSeek - High-performance models - TogetherAI - Open-source models - Fireworks AI - Fast inference - Replicate - Open-source models - xAI - Grok models - Ollama - Local model execution

Configuration¶

All configurations are created using the LlmConfig class.

OpenAI Configuration¶

import 'dotenv/config';
import { LlmConfig } from '@infinitibit_gmbh/graphbit';

const config = LlmConfig.openai({
  apiKey: process.env.OPENAI_API_KEY || '',
  model: 'gpt-4o-mini' // Optional - defaults to gpt-4o-mini
});

Azure OpenAI Configuration¶

const config = LlmConfig.azureOpenai({
  apiKey: process.env.AZURE_OPENAI_API_KEY || '',
  deploymentName: 'gpt-4o-mini',
  endpoint: 'https://your-resource.openai.azure.com',
  apiVersion: '2024-10-21' // Optional
});

Anthropic Configuration¶

const config = LlmConfig.anthropic({
  apiKey: process.env.ANTHROPIC_API_KEY || '',
  model: 'claude-3-5-sonnet-20241022'
});

MistralAI Configuration¶

[!IMPORTANT] The model parameter is required for MistralAI.

const config = LlmConfig.mistralai({
  apiKey: process.env.MISTRALAI_API_KEY || '',
  model: 'mistral-large-latest'
});

Ollama Configuration (Local)¶

[!IMPORTANT] The model parameter is required for Ollama. Make sure the model is pulled locally with ollama pull <model-name> before use.

const config = LlmConfig.ollama({
  model: 'llama2',
  baseUrl: 'http://localhost:11434' // Optional
});

OpenRouter Configuration¶

[!IMPORTANT] The model parameter is required for OpenRouter. Model format is provider/model-name.

const config = LlmConfig.openrouter({
  apiKey: process.env.OPENROUTER_API_KEY || '',
  model: 'openai/gpt-4o'
});

Other Providers¶

ByteDance ModelArk¶

const config = LlmConfig.bytedance({
  apiKey: process.env.BYTEDANCE_API_KEY || '',
  model: 'seed-1-6-250915'
});

DeepSeek¶

[!IMPORTANT] The model parameter is required for DeepSeek.

const config = LlmConfig.deepseek({
  apiKey: process.env.DEEPSEEK_API_KEY || '',
  model: 'deepseek-chat'
});

Perplexity¶

[!IMPORTANT] The model parameter is required for Perplexity.

const config = LlmConfig.perplexity({
  apiKey: process.env.PERPLEXITY_API_KEY || '',
  model: 'llama-3.1-sonar-small-128k-online'
});

TogetherAI¶

const config = LlmConfig.togetherai({
  apiKey: process.env.TOGETHER_API_KEY || '',
  model: 'openai/gpt-oss-20b'
});

Fireworks AI¶

const config = LlmConfig.fireworks({
  apiKey: process.env.FIREWORKS_API_KEY || '',
  model: 'accounts/fireworks/models/llama-v3p1-8b-instruct'
});

Replicate¶

const config = LlmConfig.replicate({
  apiKey: process.env.REPLICATE_API_KEY || '',
  model: 'meta/llama-2-70b-chat'
});

xAI (Grok)¶

const config = LlmConfig.xai({
  apiKey: process.env.XAI_API_KEY || '',
  model: 'grok-beta'
});

AI21¶

const config = LlmConfig.ai21({
  apiKey: process.env.AI21_API_KEY || '',
  model: 'jamba-1.5-large'
});

HuggingFace¶

const config = LlmConfig.huggingface({
  apiKey: process.env.HUGGINGFACE_API_KEY || '',
  model: 'meta-llama/Llama-2-7b-chat-hf'
});

Usage¶

Once configured, pass the config object to AgentBuilder or Executor.

import { AgentBuilder } from '@infinitibit_gmbh/graphbit';

const agent = await new AgentBuilder('My Agent', config)
  .build();