@kb-labs/adapters-openai

Part of KB Labs ecosystem. Works exclusively within KB Labs platform.

OpenAI language model adapter supporting GPT-4, GPT-3.5, and other OpenAI models with streaming and function calling.

Overview

| Property | Value | |----------|-------| | Implements | ILLM | | Type | core | | Requires | None | | Category | AI |

Features

• Multiple Models - GPT-4, GPT-4 Turbo, GPT-3.5 Turbo
• Streaming Support - Real-time token streaming
• Function Calling - Native tool/function support
• Configurable - Temperature, max tokens, and more

Installation

pnpm add @kb-labs/adapters-openai

Configuration

Add to your kb.config.json:

{
  "platform": {
    "adapters": {
      "llm": "@kb-labs/adapters-openai"
    },
    "adapterOptions": {
      "llm": {
        "apiKey": "${OPENAI_API_KEY}",
        "model": "gpt-4-turbo",
        "temperature": 0.7
      }
    }
  }
}

Options

| Option | Type | Default | Description | |--------|------|---------|-------------| | apiKey | string | - | OpenAI API key | | model | string | "gpt-4-turbo" | Model to use | | temperature | number | 0.7 | Sampling temperature (0.0 to 2.0) | | maxTokens | number | - | Maximum tokens to generate |

Usage

Via Platform (Recommended)

import { usePlatform } from '@kb-labs/sdk';

const platform = usePlatform();

// Simple chat
const response = await platform.llm.chat([
  { role: 'user', content: 'Hello!' }
]);

// Streaming
for await (const chunk of platform.llm.stream([
  { role: 'user', content: 'Tell me a story' }
])) {
  process.stdout.write(chunk.content);
}

// With function calling
const result = await platform.llm.chatWithTools(
  [{ role: 'user', content: 'What is the weather?' }],
  [{ name: 'getWeather', parameters: { ... } }]
);

Standalone (Testing/Development)

import { createAdapter } from '@kb-labs/adapters-openai';

const llm = createAdapter({
  apiKey: process.env.OPENAI_API_KEY,
  model: 'gpt-4-turbo'
});

const response = await llm.chat([
  { role: 'user', content: 'Hello!' }
]);

Adapter Manifest

{
  id: 'openai-llm',
  name: 'OpenAI LLM',
  version: '1.0.0',
  implements: 'ILLM',
  capabilities: {
    streaming: true,
    custom: {
      functionCalling: true,
    },
  },
}

FAQ

Change the model option:

{
  "adapterOptions": {
    "llm": {
      "model": "gpt-3.5-turbo"
    }
  }
}

The adapter includes automatic retry with exponential backoff. For high-volume usage, consider using OpenAI's usage tiers or implement request queuing.

Related Adapters

| Adapter | Use Case | |---------|----------| | @kb-labs/adapters-vibeproxy | Local multi-provider proxy (Claude, GPT, etc.) |

License

KB Public License v1.1 - KB Labs Team