@cianfrani/lm-studio

LM Studio provider for the AI SDK.

Installation

npm install @cianfrani/lmstudio

Usage

Chat API

import { lmstudio } from '@cianfrani/lmstudio';
import { generateText } from 'ai';

const { text } = await generateText({
  model: lmstudio('qwen2.5-7b-instruct'),
  prompt: 'Hello, how are you?',
});

Responses API (Beta)

The responses API provides access to reasoning traces and structured outputs from models that support reasoning.

Installation

npm install @cianfrani/lmstudio@responses

Usage

import { lmstudio } from '@cianfrani/lmstudio';
import { generateText } from 'ai';

const { text } = await generateText({
  model: lmstudio('gpt-oss', { api: 'responses', reasoningEffort: 'medium' }),
  prompt: 'Solve this complex problem...',
});

API

createLMStudio(options?)

Creates an LM Studio provider instance.

Options

• baseURL?: string - Base URL for LM Studio API. Defaults to http://localhost:1234/v1.
• apiKey?: string - API key for authentication. LM Studio typically doesn't require this.
• headers?: Record<string, string> - Custom headers to include in requests.
• fetch?: FetchFunction - Custom fetch function.
• api?: 'chat' | 'responses' - API to use. Defaults to 'chat'. (Responses API requires beta version)
• reasoningEffort?: 'low' | 'medium' | 'high' - Reasoning effort for responses API.

lmstudio

Default LM Studio provider instance.

lmstudio(modelId, settings?)

Creates a chat model instance.

lmstudio.textEmbeddingModel(modelId, settings?)

Creates an embedding model instance.

Model IDs

Chat Models

• qwen2.5-7b-instruct
• gpt-oss
• Any other model ID supported by LM Studio

Embedding Models

• text-embedding-nomic-embed-text-v1.5
• all-MiniLM-L6-v2
• Any other embedding model supported by LM Studio

License

MIT